Skip to main content

Full text of "The Problem of Loki"

See other formats


Prof. Dr. Kaarle Krohn, 




an der U^ySiVsitat Berlin 
N 24, Am Kupfergraben 7 



B Y 

Jan de Vries 







nil tam difficile est, 
quin quaerendo in- 
vestigari possit. 

In beginning the investigation of one of the most difficult 
problems of Germanic mythology it will be wise to take into 
consideration the preliminary question as to how such a 
study should be carried on. As far as we are concerned 
with literary sources — and most of our material happily 
consists in poems and prose tales belonging to the Old-Norse 
literature — a sound philological method will be the only 
advisable way to arrive at any valuable conclusion. But 
often this does not carry us very far, because our knowledge 
of the literary monuments of the different myths of Loki 
is very fragmentary and in a great many cases these are 
themselves in need of further elucidation; by following only 
the light, which they are able to give, we keep on groping 
in the dark. Scholars have therefore tried to collect more 
material from other fields of study as well: it is mainly 
etymology, popular lore and ethnology that have furnished 
hitherto the most valuable results for the study of the cha¬ 
racter of Loki. 

The meaning of the name of a god is of course a point of 
no slight importance, when we wish to grasp the original 
idea of the deity. The experience, however, gathered by 
the consideration of so many sagacious attempts to explain 
the character of a god by an interpretation of his name, may 
make us cautious and even sceptical. The divine names 

4 Jan de Vries, The problem of Loki FFC110 

are commonly very obscure, whether it be because they 
belong to a period of the language, which lies so far back, 
that etymology is no longer a safe guide, or because of the 
influence of those curious customs in the sphere of religious 
thinking, well known to scholars of ethnology as »taboo» 
and »noa». The explanation of such a name is therefore in 
most cases nothing but loose guess-work. Even in the rare 
cases, in which the etymology of a divine name is fairly 
certain, we should not think that the problem has been 
solved; the name, indeed, may belong to a very remote period 
and then the data furnished by later phases of evolution 
do not in the least agree with the character of the god 
revealed by his name. The name, however, may be late as 
well and then it only serves to elucidate a definite period in 
the development of the ideas about the deity, or it only 
reflects a peculiar side of people’s impressions of him; 
then an etymology, that in itself may be unimpeachable, 
still leads us astray, because we get a one-sided idea instead 
of a right understanding of all his different qualities. Ety¬ 
mology is not a basis to start from, but it may be a welcome 
confirmation of results arrived at by other methods of 
investigation. The resemblance of the name Loki and the 
Old-Norse word logi »fire» has induced many scholars to 
consider him as a fire-demon, an opinion which is even now 
still generally accepted. But if we try to explain the diffe¬ 
rent myths of Loki by this assumption, we have in fact 
settled the question of their meaning before examining it; a 
warning example is Elard Hugo Meyer, who in his book on 
the »Mythologie der Glermanen» rejects the connection be¬ 
tween Loki and Othin and on the other hand accepts the 
myths, where Loki is the companion of Thor, only because 
he has identified Loki and logi and therefore considers him 
to be »fire» in its movability and noxiousness 1 . 

1 L. c. p. 276 sqq. 

FFC 110 

Chapter I, Methodological prolegomena 


The scanty information, furnished by the old literary 
monuments, seems to gain valuable completion by the 
material of modern popular lore. To make serious objections 
to this method of studying problems in the field of religious 
phenomena, may seem to be a proof of unwarranted hyper¬ 
criticism, especially now-a-days, when it is so generally 
accepted. Still I wish to maintain my objections to it, 
which I have more fully developed in a paper on »Die Bedeu- 
tung der Volkskunde fur mythologische und religions- 
geschichtliche Untersuchungen» published in the Germa- 
nisch-Romanische Monatschrift (XX, p. 27—39). I may 
refer the reader to my discussion in this article. Moreover I 
have urged the same opinion in my Contributions to the 
Study of Othin especially in his relation to agricultural prac¬ 
tices in modern popular lore», published in the FFCommuni- 
cations No 94; here I had the opportunity to show that the 
character of a god, as it appears in the popular tradition 
of modern times may be entirely different from the original 
meaning of the deity in the heathen period. On the other 
hand I have also endeavoured to show the strong difference 
between an old religious conception and modern popular 
representations in an article »Van Alven en Elfen», that 
appeared in the Nederlandsch Tijdschrift voor Volkskunde \ 
So I do not consider it necessary to repeat my objections 
to a method, which, tempting as it may be in cases where 
genuine heathen tradition is too fragmentary or too obscure 
to render us valuable material for our investigation, still 
is highly questionable when we want to understand the 

1 Ibid. XXXVI, p. 3—30. Cf also my articles »Het huidige bijge- 
loof als bron voor de kennis van den heidenschen godsdienst der 
Germanen» in the same periodical XXXVII p. 25—35 and »Volks- 
kunde en Volkenkunde als hulpwetenschappen der godsdienstge- 
schiedenis in the Dutch periodical Mensch en Maatschappij VIII 
(1932) p. 452—464. 


Jan de Vries, The problem of Loki 

FFC 110 

meaning of a religious phenomenon in its original form. The 
present study will supply fresh evidence for my opinion. 

If we consider a religious phenomenon as a part of a 
definite system of belief, it may often seem very difficult 
to understand its meaning as a particular form of the reli¬ 
gious representations: a comparison with corresponding 
phenomena among other peoples may in such a case offer 
valuable suggestions for its understanding. These peoples 
may live in a more or less close neighbourhood, or their 
civilisation may be supposed to have exerted a certain in¬ 
fluence upon the tribe whose religious belief we are studying; 
in this case the idea easily presents itself, that a resemblance 
between the religious representations must be attributed to 
the fact that one borrowed from the other. If, on the con¬ 
trary, the peoples belong to quite distinct geographical 
and cultural spheres, we must conclude that corresponding 
religious forms are the result of corresponding religious 

In the latter case it may appear that we have the oppor¬ 
tunity to penetrate into the core of a religious phenomenon, 
because it now seems free from all secondary local and temp¬ 
oral accretions and accessible to an inquiry into its real 
character. Since I shall avail myself of the opportunity to 
explain the meaning of Loki by adducing fresh evidence 
from so-called primitive peoples, I can not but approve of 
this method. I wish, however, to emphasize the danger 
which lies in such a comparison. A religious phenomenon 
is the outcome of a very complicated development; it is 
intimately connected with the way of thinking and feeling 
of man and dependant upon the special cultural sphere, in 
which it has grown up. So there may even be a strong 
resemblance in outward form and nonetheless a striking 
difference in the real religious meaning between the corres¬ 
ponding representations of two peoples. There is moreover 
the possibility that, by such a comparison with the beliefs 

FFC 110 Chapter I, Methodological prolegomena 


of primitive peoples, we reduce the importance of a reli¬ 
gious representation of a higher cultural level to its minimum, 
that is to the simple ideas that underlie the later develop¬ 
ment. Things may seem to be the same and really be quite 
different. Lastly we must bear in mind, that the material 
of the so-called primitives must be used with extreme cau¬ 
tion, as it is not always collected in a sufficiently scientific 
and trustworthy manner. The critical evaluation of the 
sources must precede their use, but it is not always possible 
to do so when we enter on a field of scientific research as 
mere visitors. We had better stay within our own frontiers. 
A religious phenomenon can only be rightly understood, if 
we are able to consider it as part and parcel of the whole 
system of belief; but when we use the material of primitive 
peoples for the elucidation of a religious phenomenon of a 
higher order, we run the risk of detaching it from its basis 
and of considering it as a phenomenon, independant of 
time and space. 

The danger of misinterpretation seems much less if 
we confine ourselves to the comparison with the beliefs of 
peoples, who belong to the same linguistic group or stand on 
the same cultural level. Old-Norse mythology is naturally 
connected with the beliefs of other Indo-European peoples; 
the highly developed mythologies of the Aryans, the Greeks 
and the Romans may yield many valuable suggestions for 
explanation. If, however, we find common ideas in these 
religions, how must we consider them? As a proof of the 
common stock of religious representations lying at the bot¬ 
tom of the different mythologies that developed quite 
independantly? Or as the result of mutual borrowing, in 
which case the Old-Norse tradition as the later one is likely 
to be considered as the debtor to the other Aryan religions? 

It is a well-known fact that after the highly conjectural 
hypotheses which the age of Romanticism built in the field 
of Germanic and Aryan mythology, a reaction has taken 


Jan de Vries, The problem of Loki 

FFC 110 

place tending to explain congruities between the religious 
representations of kindred peoples as borrowings by the 
lower civilised from the more advanced. It is a fact, that by 
some peoples a great cultural influence has been exerted on 
their neighbours. But on the whole we must insist upon the 
necessity of a protracted and intimate connection by such 
peoples to allow religious beliefs and customs to wander 
from one country to an other. I think it important to 
enphasize this point, since several scholars have tried to 
explain various peculiarities in the myths of Loki by such 
farreaching influences, As for the r61e of Christianity in the 
development of the Old-Norse mythology, it can not be 
denied of course that a considerable part of the myths we 
know of have been created or at any rate have acquired 
their literary form during the period when the Scandinavian 
peoples were deeply impressed by Christian ideas. Still 
we ought to be cautious not to overestimate this influence 
and to keep in mind the possibility that a myth in its ap¬ 
pearance may be altogether Christian and nonetheless have 
a quite heathen origin. 

If we suppose an Old-Norse myth to be related to tradir 
tions of peoples living at a great distance from Scandinavia, 
e.g. to those of Asia Minor, we must not content ourselves 
by pointing out a series of similarities, however convincing 
they may seem to be, but we are obliged to indicate the way 
by which the myth has travelled and the circumstances which 
make such a transmission plausible. Casual merchants or 
vikings, going to and fro between distant countries can 
not be made responsible for the transplantation of a myth, 
unless it is nothing but a curious tale dressed up in the 
hieratic robes of a full-blown myth. For in this case the myth 
belongs exclusively to the domain of literature and has to 
be studied as any other, popular tale x . 

1 I do not intend to deny the literary character of autochthonic 

FFC 110 

Chapter I, Methodological prolegomena 


The myths of Loki are, to be sure, for a considerable part, 
of this latter kind. This circumstance makes it rather easy 
to arrive at a conclusion, for by the method of the modern 
study of popular traditions we may be able to point out 
the dependance of such an Old-Norse myth upon a widely 
spread folk-tale. On the other hand this character of the 
majority of the traditions about Loki hampers the right 
understanding of the deity himself, because most of our 
material proves to be of no use for the study of the religious 
value of Loki. Hence we must not be too radical in cutting 
away from the bulk of traditions all those tales which are 
under the suspicion of being only adaptations of ordinary 
folk-tales; otherwise at the end of our work the result may 
be that of the swollen stream of Loki myths a dry bed is 
the only trace left. 

Our task will be a critical survey of all the available 
material and as this belongs exclusively to Scandinavian 
literature, we must begin with a philological investigation. 
A folkloristic study comes next to separate the chaff 
from the wheat. If possible we will make a distinction 
between the later and the earlier strata of the tradition, to 
guard against mixing up the representations of different 
ages. It should not be our principal aim to find the oldest 
form of the Loki-mythology, but to arrive at a coherent 
and intelligible group of myths, by which we are able to 
understand the meaning of Loki in a definite period. How¬ 
ever tempting the construction of an evolutionary scheme 
may be, we must always remember that a religious pheno¬ 
menon is not sufficiently explained when we know where 
it came from, but only when we understand what it really is. 

mythological tales as well — Haggarty Krappe, The Science of 
Folklore p. 317—318 has rightly insisted upon it — but here we have 
always to reckon with the possibility that there is an underlying 
basis of actual belief. 



In his monumental work on Teutonic mythology Jacob 
Grimm , 1 has first pointed out the close affinity of the words 
Logi and LoTci which he can not accept to be fortuitous: 
while Logi is »fire» conceived as a natural force, Loki re¬ 
presents a more advanced stage of this idea and may be com¬ 
pared at the same time with the Greek deities Prometheus 
and Hephaistos. If he has been known once in Germany no 
traces of him are now left; in Scandinavia however several 
popular traditions about a demon Loki or Lokke are still in 
existence. Furthermore Grimm considers the possibility 
of the word Loki, which indeed was originally a form of 
Logi altered according to the German soundshifting 2 , 
being brought into connection with the root lukan »claudere»; 
then the name of this god got the meaning of »the closer» 
and he may be compared with the anglosaxon demon Gren- 
del and the German word hdllriegel. Such a confusion of 
mythical notions is a proof of the high antiquity of the deve¬ 
lopment of Logi to Loki. 

This theory has been subsequently accepted by a great 
number of scholars, who only slightly modified or else more 
clearly formulated the rather hazy conceptions of Grimm. 
The Danish scholar N. M. Petersen expresses as his view 
that Loki is a fire-god, and he adds, that he belongs to hea¬ 
ven by his affinity with Othin, to earth as a Fenrisulfr and 

1 Deutsche Mythologie, 4th edition, I, p. 199 sqq. 

2 Grimm says not quite clearly: das im Laut fortgeschobne Loki. 

FFC 110 Chapter II, Critical survey of the extant theories 


to the underworld as the fettered demon x . Less homogene¬ 
ous is the conception of Th. Wisen 1 2 , who accepts the expla¬ 
nation of Loki as a fire-god, as far as the name is concerned; 
he was once the same as Logi, but since these religious perso¬ 
nalities were separated, his name had to undergo a slight 
alteration. So the form Loki is, according to Wisen, due to a 
deliberate changing of the name! From the Old-Norse 
poems other qualities of this god may he concluded: he is a 
deity of the water, because Fenrir seems to be connected 
with the word fen »moor»; he is also a god of the air, as is 
proved by his surname Loptr and a couple of mythological 
features. The study of Wisen is of course quite antiquated, 
in method as well as in results, but there is one observation 
that in course of time proved to be very useful for further 
investigation: in popular tradition an older form of this 
deity may have been preserved than in the Eddie poems of 
the last period. 

The resemblance between Loki and Prometheus, which 
indeed can not be denied, was mostly considered to be a 
proof of his character as a firegod, even going back to the 
Aryan period. The sagacious Swedish scholar V. Rydberg 
argued in the same way, considering him only more parti¬ 
cularly to be connected with the heavenly fire, the lightning; 
this seems to be shown by the etymological meaning of the 
names Byleistr and Farbauti both parents of Loki 3 . The 
more specialised idea of Loki as a demon of the lightning 
seemed to be generally accepted. R. Much called to mind his 
connection with Thor in a great many mythological tales, 
but at the same time drew our attention to his character of 

1 Nordisk Mytologi (Swedish translation), 1869, p. 309. 

2 Oden och Loke (1873) p. 62 sqq. 

3 Undersokningar i Germanisk Mythologi (1886) I, p. 450 sqq; 
Byleistr is explained with the help of the words bylr »whirlwind» 
and eistr »who lives in the East». Farbauti means »who stakes people 
with perdition*. 


Jan de Vries, The problem of Loki 

FFC 110 

heavenly smith, by which his relation to Velend and Vadi 
seemed to gain a plausible explanation l . As to the name, he 
rejects the connection with Logi as well as Bngge’s theory 
about its being derived from the word Lucifer; he thinks 
it worth consideration that Loki may correspond with ags 
hca »prison», and ventures to conclude that he is the repre¬ 
sentative of the realm of death as he was the first and the 
most prominent one to be cast into it. One should imagine 
that he would accept the hypothesis of S. Bugge, that Loki 
was a northern development of the Christian devil, but he 
shrinks from doing so, and is even of opinion that this may 
be the result of an independant evolution of the Teutonic 

Much rejects emphatically the confused explanation 
given by E. Mogk 2 , whose theory I will treat further on. 
The handbooks of P. Herrmann and E. H. Meyer, published 
in the same year, 1903, do not bring us any farther; the for¬ 
mer repeats the conception of Loki as a fire-demon, as 
proved by his name, his mythological relatives, his art of 
forging and popular tradition; he adds moreover that he 
became in course of time more ethically determined, without 
however adducing any evidence for the reasons of this 
curious development 3 . Still more confused is the opinion 
of E. H. Meyer 4 , who starting from the conception of 
Loki as a fire-god, makes this very puzzling definition: 
Loki is the lightning, then also the heat of summer and the 
air vibrating on a hot summer-day, furthermore the fire on 
the hearth and lastly the vulcanic earth-fire. It need not be 
said that such constructions, emanating from the writing- 

1 Der germanische Himmelsgott (Festschrift Paul, 1898) p. 48sqq. 

2 In his treatment of the Germanic mythology in Paul’s Grundriss 
der germanischen Philologie, repeated in the second edition (1900) 
III, p. 347. 

3 Nordische Mythologie, p. 403 sqq. 

4 Germanische Mythologie p. 275 sq. 

FFC 110 Chapter II, Critical survey of the extant theories 


desk, are of no value whatever, since they do not take into 
consideration the actual religious representations of the 
heathen Skandinavians. The German scholar R. M. Meyer 
is also of opinion that Loki simply is a god of fire and explains 
by the double character of this element the contradictions 
in his attitude towards the gods. 1 

In the definition of E. H. Meyer we find the different 
aspects of Loki awkwardly combined, as he appears in Old- 
Norse mythology and in modern popular lore. In fact this 
deity shows so many different features, that it seems 
impossible to unite them into one single original idea. So 
gradually it became clear, that Loki’s character, as it is 
shown by Eddie and Skaldic poetry, is the result of a long 
and complicated evolution, influenced by distinct religious 
representations; consequently it may be possible too that 
different mythical beings have coalesced into the incongruous 
figure of Loki. The fire-god is then only one of its aspects, 
predominating or less important according to the fancy of 
the scholars. Since the strong resemblance between him and 
Prometheus remains an undeniable fact, there is always a 
strong tendency to explain this likeness by the theory of an 
Aryan fire-deity. So G. Wilke combines the Scandinavian 
Loki with the fire god whom Caesar mentions as one of the 
chief-deities of the Germanic tribes and considers the cor¬ 
responding gods Loki, Prometheus, Vulcan and Agni-Matar- 
igvan as proof of the existence of a common Indo-European 
fire-god. 2 

The etymology, that explains the name Loki by a verbal 
root of the meaning »to close», has since J. Grimm become 
the basis of a second hypothesis about the original meaning 
of this god. We find it fully developed in Goliher 's handbook 3 

1 Altgermanische Religionsgeschichte (Leipzig 1910) p. 337. 

2 Die Religion der Indogermanen in archaologischer Beleuchtung 
(Mannus-Bibliothek Nr. XXXI) p. 119. 

3 Handbuch der germanischen Mythologie (1895) p. 406 sqq. 


Jan de Vries, The problem of Loki 

FFC 110 

who explains him as »the closing god, the god who causes 
the world’s end». His development is considered, clearly in 
accordance with S. Bugge’s views, as due to the influence 
of the Christian Lucifer. The character of a fire-god is to be 
attributed to the fact, that a fire-demon Logi has been fused 
with Loki, not only because there is a strong resemblance 
between the names, but also because the world is finally 
destroyed by fire. 1 Still more abstract is the explanation 
given by E. Mogk 2 , who maintains the etymology of Loki 
by means of the verb lukan, but translates it as the »god 
who makes an end to the agreeable and to the disagreeable» 
and consequently the friend as well as the foe of the gods. 
It is in fact not difficult to adduce examples of Loki’s 
ambivalent character from the Old-Norse mythology; it is, 
however, a rash conclusion to say: »This only possible ety¬ 
mology of the word shows, that Loki belongs to a late 
period of the creation of myths, to a time when abstract 
ideas were drawn into the sphere of mythological poetry 
and further developed.» We might be astonished at finding 
such a lack of understanding with regard to religious phe¬ 
nomena in a scholar, who professes to give a treatment of the 

1 As a curious example of the philosophical constructions of 
German mythologists this sentence of Golther deserves to be quoted 
in its original form: Ein Damon des Feuers, welcher auch sonst unter 
dem Namen Logi, die personlich gewordene Lohe vorkommt, scheint 
vielleicht wegen des anklingenden Namens und, da ja wirklich die 
Welt im Feuer endigt (logi — loki: Lohe — Endiger), mit Loki ver- 
schmolzen zu sein. 

2 In the Grundriss III, p. 347. Many years later Mogk abandoned 
this theory and accepted the explanation of Loki as a fire-demon 
(Feueralf). Cf. his article in Hoops’s Reallexikon der germanischen 
Altertumskunde III, p. 162 sqq. He concludes with the prudent 
remark: In the poetry about Loki nothing of a unity is to be seen and 
consequently it is impossible to point out one single fundamental 
trait in his appearance (. . . und daher lasst sich auch im Wesen der 
Erscheinung kein einheitlicher Grundton festlegen). 

FFC 110 Chapter II, Critical survey of the extant theories 


mythology of the Teutons, if we did not remember that 
the problems of Germanic religion were mostly considered as 
belonging to the field of philology and accordingly that a 
Skandinavian myth had to he explained as any other product 
of poetic fancy. So it is quite intelligible that a scholar like 
Chantepie de la Saussaye should hesitate before such a 
conclusion, but although he rejects the explanation of Loki 
as a fire-god and at the same time as a »closer » 1 , he does not 
give any better hypothesis in its stead. He contents himself 
with quoting Mogk’s definition: »the true impersonation of a 
fTmlrwho takes delight in snapping his fingers at the company 
round about him, but who always knows how to escape the 
net that is spread for him». But this applies only to the last 
stage of the development of the god, more particularly of 
Loki as he appears in a couple of Scandinavian myths and 
in the Lokasenna. 

The same opinion slightly modified is enunciated by 
F. Jonsson 2 , who accepts the explanation of the name as 
»the closer» and considers him as belonging to the giants and 
consequently elected to do all possible mischief in the world. 
That is the reason why he entices Othin to engage in a bond 
of bloodfraternity and why he seduces the goddesses to 
adultery. He is the personification of all that is bad and 
devilish. He has nothing whatever to do with nature nor 
with fire. As we might expect from a scholar like F. Jonsson, 
this theory is clear and decisive, but it is not quite in accor¬ 
dance with the facts, presented by the tradition, nor does 
it take into consideration the possibility that the figure of 
Loki also belongs to the domain of religious belief. 

The view of 0. Schoning 3 , that Loki on the ground of his 

1 The Religion of the Teutons (1902) p. 260 sqq. 

2 Godafrae&i Nor&manna og Islendinga eftir heimildum (Reykja¬ 
vik 1913) p. 96. 

8 Dodsriger i nordisk hedentro (Copenhagen 1903). Loki’s 
eating contest with Logi in the hall of Utgarhaloki is adduced as a 
proof for his being a devourer of corpses! 


Jan de Vries, The problem of Loki 

FFC 110 

connection with Hel,has been originally a god of the dead, has 
not found many adherents. The Swedish scholar H. Schuck 1 
considers him also as a chthonic god, a lord of the realm of 
death, an opinion which he afterwards under the impression 
of Axel Olrik’s investigations modified in this way by 
explaining the Icelandic Loki as a combination of an original 
death-god and a servant of the thundergod. I did not find 
any further indication about the way in which this curious 
religious development might have taken place. 2 

Deities of death usually show some connection with the 
conceptions of fertility: so, if it were well established that 
Loki originally belongs to the realm of the dead, we might 
expect to find some traces of his character as a god of 
vegetation. F. R. Schroder advocates this opinion in his 
treatment of the Germanic religion in Clemen’s handbook 
»Die Religionen der Erde» 3 , but as he only expresses it 
hesitatingly and adduces no proofs for his view, we are 
unable to criticise its value. 

The resemblance between Loki and Prometheus was 
explained by the scholars as a proof of their both being 
fire-gods; but there was still an other way of understanding 
the remarkable fact that two supernatural beings, although 
belonging to quite distinct mythological systems and in fact 
showing several important differences, nonetheless in some 
respects appear in the same functions. They are both typical 
forms of the culture-hero. To this side of their character the 
eyes of scholars were first opened when ethnology had col¬ 
lected material enough to give an idea of the primitive forms 

1 Studier i Nordisk Litteratur- och Religionshistoria II, p. 125. 

2 In Schiick-Warburg, Illustrerad Svensk Litteraturhistoria 
(3d edition) I, p. 159. 

3 He speaks about Loki in one single sentence that runs as follows: 
»So diirfte ein Wachstumsgott ursprunglich auch Loke gewesen 
sein, der in seiner spateren Entwicklung vielleicht einzelne Ztige 
des christlichen Teufels angenommen hat.. .». 

FFC 110 Chapter II, Critical survey of the extant theories 


of this well-nigh universally known religious conception. 
It was an ethnologist who first pointed out the strong 
resemblance between Loki and the culture-hero of the Algon¬ 
quin tribes, but Leland made the absurd conclusion that 
this was due to the cultural influence of the Norsemen upon 
the North-American Indians. 1 This observation remained 
unknown to the European scholars, who studied Germanic 
mythology; not until a quarter of a century later did F. von 
der Leyen explain the curious myths of Loki by his funda¬ 
mental character of a culture-hero. 2 He was originally a real 
benefactor of gods and men; he played his tricks only to 
obtain all kinds of indispensable things from malevolent 
demoniac beings. In course of time Loki was confused with 
the Christian devil; in Iceland, and only there, was he mixed 
up with a fire-demon Logi. Prom Old-Norse literature we 
get a wrong impression of this god, for the poets have 
changed him into the unrelenting foe of the cosmic order. 

This keen understanding of the puzzling problem of Loki 
has not had the effect it deserved. It has, of course, not 
been denied that Loki in some myths plays the role of the 
culture-hero, but this character has not been considered as 
essentially his own. Starting from the opinion that Loki, 
as he appears in the Old-Norse myths is a very complicated 
being, the type of the culture-hero was allowed to explain one 
side of his multivarious character. Axel Olrik 3 , who accepts 
v. d. Leyen’s explanation, agrees to this as does also Mac 
Culloch, who however emphasizes more strongly the character 

1 C. G. Leland, The Algonquin Legends of New England or 
Myths and Folk-Lore of the Micmac, Passamaquoddy and Penobscot 
Tribes (London 1884). The resemblance between Loki and the 
culture-hero of the Algonquins has also been observed by Mrs E. R. 
Emerson, Indian Myths (Boston 1884) p. 361—3. 

2 Die Gotter und Gottersagen der Germanen (Deutsches Sagen- 
buch I, Miinchen 1909) p. 222 sq. 

3 In his paper »Myterne om Loke»in Festskrift Feilbergp. 573 sqq. 



Jan de Vries, The problem of Loki 

FFC 110 

of the mischief-maker, often the concomitant feature of the 
culture-hero. 1 Since I intend to discuss this hypothesis 
in the course of this paper at full length, I will here refrain 
from further argument. 

Hitherto the hooks, we have taken into consideration, 
have been for the most part general surveys of Germanic or 
Scandinavian mythology; now we must turn to the most 
valuable of the scientific literature about Loki, those papers 
which deal exclusively with this deity. Here we shall meet 
the names of Axel Olrik and Hilding Celander. Olrik began 
his investigations with several smaller articles in the perio¬ 
dical »Danske Studier». In 1905 he published a paper 
entitled »Tordenguden og hans dreng» 2 in which he compared 
an Bsthonian myth of the son of the thundergod with the 
Eddie poems HymiskviSa and J>rymskvi8a; he concludes 
that Loki has been the son or the servant of the thundergod 
and that he may be consequently compared with the figure 
of Jj&lfi in Scandinavian mythology. He tries to explain 
the fact that the same religious conception is represented by 
two different mythological beings in this way that they 
belong to different geographical areas, bjalfi belonging espe¬ 
cially to the Swedish people, Loki as the thunderer’s son to 
the countries lying east of the Baltic sea. The value of the 
Esthonian material will be discussed later on; it may suffice 
here to draw attention to the great improbability of Olrik’s 
conclusion: in fact considering the relations of the Swedes 
with Finland and Esthonia in heathen times, we might 
expect that in these latter countries Swedish religious ideas 
would prevail. According to Olrik the contrary is the case; 
in these regions we not only find a different type of this 
servant of the thundergod, but moreover traces of him are 
rather unexpectedly left in the West-Scandinavian mytholo¬ 
gical traditions. 

1 Eddie Mythology (Boston 1930) p. 149. 

2 Ibid. p. 129—146. 

FFC 110 Chapter II, Critical survey of the extant theories 


Olrik, however, was convinced that he had taken hold 
of one thread of the intricate tissue of Loki myths. The 
next step was taken, as might be expected from a scholar 
like him, on the field of popular tradition. »Loki i nyere 
folkeoverlevering» is the title of a couple of articles, published 
in 1908 and 1909, where he collects all the available material 
from the Faroes, the Shetland-Islands, Denmark, Sweden and 
Norway. 1 He arrives at the conclusion that according to 
popular lore Loki is a small goblin, mainly connected with 
fire and other related phenomena, whose character of liar, 
trickster and thief well agrees with the image the Old-Norse 
literature draws of him. 

The popular traditions were studied meanwhile by the 
Swedish folklorist Hilding Celander, who criticized in many 
respects the conclusions of Olrik 2 and then presented as his 
own opinion, that Loki was originally a spirit, of the kind 
Scandinavian peoples call bergmand , nisse or vdtte , and that 
he was intimately connected with the veneration of the 
dead. Of course he finds in the literary tradition many details, 
which may be used as proof of his theory; on the other hand 
there are also several myths that do not agree at all; they 
are to be explained by secondary influences. These are 
partly nothing but ordinary motives of folk-tales (such as 
the myths of I8unn, jazi and Ska5i), they are due partly 
to traditions about the fettered demon, the importance of 
which had been pointed out by Axel Olrik in two brilliant 
studies about the myth of Ragnarok. 3 The connection of 
Loki with Hel and the myth of his being bound by the gods 
are explained in the latter way. The view, that Loki was a 

1 Danske Studier 1908, p. 193—207 and 1909, p. 69—84. 

2 Lokes mytiska ursprung in Sprakvetenskapliga Sallskapets i 
Uppsala forhandlingar 1906—1909 (Uppsala 1911) p. 18—140. 

3 Ora Ragnarok in the Aarboger for Nordisk Oldkyndighed og 
Historie 1902 p. 157—291 and Ragnarokforestillingernes Udspring 
in Danske Studier 1913, p. 1—283. 


Jan de Vries, The problem of Loki 

FFC 110 

culture-hero is rejected on the consideration that he never 
properly steals cultural goods. 1 The curious myth of the 
close connection between Othin and Loki is shortly dismissed 
by the observation that they both belong to the realm of 
death. In his opinion Loki has lost most of his importance 
in the myths we are acquainted with. So he thinks it possible 
that a considerable time before the tenth century the 
development of an elf-like spirit into a god-like being, 
comparable with the highest of the gods, had taken place, 
and that the Old-Norse myths show only the corrupted and 
obliterated features of this deity, while on the other hand 
more than ten centuries later we find in popular lore the 
traces of the original spirit, which the heathen Scandinav¬ 
ians themselves have probably no longer known. The study 
of Celander is valuable as a careful collection of folldoristic 
material, but his method is insufficiant for the study of a 
complicated religious phenomenon as Loki. I shall have the 
opportunity to discuss the conclusions of Celander, when 
I myself study the modern traditions about Loki. Here it 
may suffice to quote the criticism of Meckel, who judiciously 
remarks that Celander treats an ideal form of Loki, which 
never has existed anywhere but in the brains of the ancient 
and the modern mythologists. 2 The positive value of Celan- 
der’s book is in fact small; of more importance is his rejection 
of Olrik’s conclusions, which holds good even against the 
new and detailed investigation, this Danish scholar soon 
afterwards published. 

This study »Myterne om Loke», which forms one of the 
most brilliant contributions to the »Festskrift Feilberg», 
makes a distinction between different mythical beings, that 
have subsequently coalesced into Loki. The Scandinavian 
myths show us on the one hand a Loki in close connection 

1 L. c. p. 116 and later again Danske Studier 1914 p. 84. 

2 Anzeiger fur deutsches Altertum XXXVI (1913) p. 135—139. 

FFC 110 Chapter II, Critical survey of the extant teories 


with Othin, on the other hand a second Loki always com¬ 
bined with Thor. The latter belongs mainly to the Eastern, 
the former to the Western parts of the Scandinavian territory. 
The malevolent Loki, although showing some traits in com¬ 
mon with both the Othin-Loki and the Thor-Loki, is created 
among the Goths in the Migration-period and came thence 
to the Northern peoples. Finally we have to consider the 
type of the cultnre-hero, probably the oldest feature of Loki; 
by means of a rather subjective reasoning he finds traces 
of this being in the Rhine-tracts as early as the first centuries 
of our era. For the Loki of modern popular belief there seems 
to be hardly any place in the evolutionary scheme of Olrik; 
he contends however that this Loki belongs to an old 
conception and may be traceable in his smallness and the 
trickster-qualities of the Eddie Loki. The hypothesis seems 
to me not less of a highly constructive character than those 
condemned by Olrik as being the outcome of an a priori 
argument; it is, moreover, very complicated and I consider 
it to be a serious shortcoming, that we must accept several 
independant mythological conceptions at the base of an 
evolution, and that we do not understand the reason, why 
these different features in course of time have been united 
in the god, the Old-Norse traditions trace for us. 

In 1931 a new discussion of the Loki-problem has been 
the subject of a Dutch dissertation. As the author, Miss 
E. J. Gras, already clearly avows by the title of her book, 
she intends to give only an account of the Old-Norse myths 
and their mutual relationship. 1 Her conclusion is the follow¬ 
ing: the tales about the devilish Loki are of no value for 
understanding his original character. The other myths are so 
encumbered with folkloristic motives, that it is very difficult 
to use them for the study of this religious phenomenon; still 

1 De Noordse Loki-mythen in hun onderling verband (Haarlem 


Jan de Vries, The problem of Loki 

FFC 110 

it appears that he was considered to be a cunning demon and 
a trickster. His connection with Othin must be very old; his 
relation to Thor, however, is due to a literary creation; so 
she does not adhere to Olrik’s theory about a Loki as the 
servant of the thundergod. He has no connection at all 
with the element of »fire»; there is more evidence for his 
relation to water. Here Miss Gras adduces the testimony 
of the Dutch water-spirit »Kludde» or »Lodder», already 
observed by Olrik 1 and consequently considers him to be a 
mythological being, belonging to the West-Germanic tribes 
as well as to the Scandinavians. The inscription on the 
brooch of Nordendorf, where we read the word Logapore is 
adduced as fresh evidence for his existence in Germany. 
From what follows it will be shown that I think these 
conclusions unacceptable, but the detailed and thorough 
discussion of the Old-Norse myths is a valuable piece of 
work, that will free me from the necessity of treating them 
at full length since I am able to refer the reader to this book. 

1 Griiner-Nielsen and Olrik, Loeke, Lodder i flamsk folketro in 
Danske Studier 1912 p. 87—90. 



Besides the modern popular tradition the main sources 
for the myths of Loki are literary monuments, belonging to 
the Old-Norse literature. As early as the beginning of the 
tenth century we find traces of a fully developed mythology 
of Loki in the poems of pjobolfr or Hvini; by his poetical 
metaphors in Ynglingatal and still more extensively in the 
Haustlong, where the skald describes a shield, on which 
several myths were painted, bjodolfr furnishes us with 
valuable material for the knowledge of the figure of Loki 
in those days. Since this poet is a Norwegian, the image he 
draws of Loki is not properly Icelandic, but must be attributed 
to the original tradition of Norway. Although some influence 
of the Christian belief is not improbable, on the other hand 
we may assume that the Norwegians about the year 900 were 
still on the whole heathen and the time seems to be insufficiant 
for such a deeplying influence of Christian ideas and legends 
to have altered the pagan myths. 

In the latter half of the same century lived the Icelandic 
skald tJlfr Uggason, who composed the famous Husdrapa 
about the beautiful hall, newly erected by Olafr pai at 
HjarSarholt and especially about the woodcarvings that 
covered its walls. According to this poem the artist had 
chosen mythological themes and that is the reason why 
TJlfr tells about the fight between Heimdallr and Loki, the 
conflict between Thor and the Midgardserpent and the 
myth of Balder. In this case evidence of Christian influence 


Jan de Vries, The problem of Loki 

FFC 110 

may be safely accepted, at all events with regard to the 
builder of the house and his family; this does not imply 
however, that these myths must have necessarily contained 
Christian elements, for when an artist makes a representation 
of a tale, he of course supposes the contents to be generally 
known. He can not carve out on the walls every detail of the 
myth, but he must select those that are of the highest 
importance and are at the same time sufficient to express 
the proper meaning of what he wants to represent. I think 
it improbable that the myths, carved on Olafr’s hall were 
new-fangled; on the contrary they reflected the popular 
tradition, current in his time. Moreover, if we are to rely 
on the testimony of this fragmentary poem, we must conclude 
that the choice of these myths was not fortuitous; they were 
for one reason or another considered to be of such an 
importance that Olafr wished to have them carved on the 
walls. The conclusion must be: these myths were well-known 
and even belonged to the common stock of these religious 
traditions, or else Olafr pai had some particular reason for 
choosing them to the exclusion of others, in which case we 
must try to find the peculiar element, that united them in 
the predilection of Olafr. 

About the year 1000 the Icelandic poet Eilifr Godrunar- 
son composed a poem about Thor’s journey to Geirrodr. 
Since it originated in a time that Christian influence had 
become allpervading in Iceland and the skald himself some 
years afterwards was converted, direct or indirect connection 
with Christian legends or at any rate a free handling of a 
myth that belonged to the realm of literature, and no more 
to that of real belief, may be easily accepted. 

Besides an occasional kenning in a skaldic poem, these 
are the principal testimonies for the myths of Loki afforded 
by the skalds. More material is to be found in the Eddie 
poems, but here we meet with the difficulty, that these 
monuments can not be dated with any certainty. Centuries 

FFC 110 Chapter III, Chronological survey of the sources 


may lie between different poems of the same Edda-collection. 
I think we may safely contend, that HymiskviSa and parts 
of Hyndluljod belong to the tenth century, that the Voluspa 
was composed in the period of transition from heathen to 
Christian belief, i. e. about the year 1000; that furthermore 
the Lokasenna has originated in a time when the heathen 
myths were still well-known, although the veneration for 
the ancient gods was nearly gone, and that this poem 
consequently was composed in the first half of the 11th 
century. As to the J>rymskvi5a opinions are very much 
divided, some dating it rather early and at any rate in the 
heathen period, others considering it on the contrary a 
creation of later extraction. I have given my reasons 
elsewhere 1 , why I accept the latter view and I will accordingly 
treat this poem as a testimony from the end of the 12th 
century; still I will take into consideration the possibility of 
its belonging to the oldest layer of Eddie poems as well. 

Finally an abundant harvest may be gleaned in the last 
period of the classical Old-Norse literature, i. e. in the 
beginning of the 13th century. Very important is the 
information given by Snorri Sturluson in his socalled Edda, 
because he not only completes the often very fragmentary 
and obscure renderings of the skalds, but also furnishes us 
with fresh evidence by giving the contents of otherwise 
unknown myths. It has often been said and it need not be 
repeated, that we must handle this material with the utmost 
caution. Snorri himself is of course a Christian and the 
myths handed down to him, have lived about two centuries 
in the memory of a thoroughly Christianized people. The 
proper meaning of the myth may have been obliterated, its 
contents modified, the details misunderstood and confused. 
Moreover, since the poets remained in the habit of using 

1 »Over de dateering der l’rymskvi5a» in het Tijdschrift voor 
Nederlandsche Taal- en Letterkunde XLVII (1928) p. 251—322. 


Jan de Vries, The problem of Loki 

FFC 110 

mythological names for their »kennings» the act of myth¬ 
making survived even in the Christian period and so it is 
not impossible, that new myths, products of a queer anti¬ 
quated literature, evolved. Finally Snorri himself is too great 
an artist to leave the raw material, he succeeded in collecting, 
quite intact; on the contrary, we could expect him to join 
the »disjecta membra» of a tale together and still suspect 
that he did not always catch the right meaning of the 
original. He may even have had its own ideas about how it 
should have been and consequently have remodelled the 
myths according to his own opinion. At any rate, the 
explanation offered by Snorri must be accepted with reserve; 
the combination of several myths must be investigated with a 
very critical eye, but the myths themselves may be accepted 
as having been current in the same form as he relates them. 

Still later is the Soria fattr; it belongs to a highly epigonic 
age. The myth of the stealing of the Brlsingamen, told by an 
author, who only wishes to delect his readers, deserves but 
slight belief, and scholars, who utterly disagree on most 
parts of the Loki-problem, are in perfect harmony about the 
untrustworthiness of this document. 

With the Faroese ballads »Lokka tattur» and »Risin og 
Lokki»we arrive at the border of modern folkloristic material. 
We may assume, however, that these poems reach conside¬ 
rably far back, because it is difficult to conceive how else a 
ballad-poet of the 16th or 17th century got the idea of 
making a poem about heathen deities. If, however, the 
study of the texts themselves leads us to the conclusion that 
they contain no genuine pagan tradition, it will not be 
difficult to explain this singular fact; the ballad literature 
of the Faroes proves that these poets were thoroughly 
acquainted with the Old-Norse traditions and that the 
people itself showed an astonishing interest in all that belongs 
to the sagas. So one of these poets may have got the idea, 

FFC 110 Chapter III, Chronological survey of the sources 


when reading Snorri’s treatise on Icelandic mythology, of 
composing a ballad about these heathen gods. 

Finally we must mention the popular lore of our own 
time. As might be expected these modern traditions about 
Lola are not very numerous; we might even wonder, that 
there are really any traditions at all, because, as far as 
we can see, Loki was in the heathen period a deity, well 
known in mythological tales, but not venerated in a public 
or private cult. Hence the question may be put: do the 
modern Loki-traditions root in genuine heathen belief and 
practice, and if so, we must try to understand the relation 
between this modern demoniac being and the heathen deity, 
who show a different character in almost every respect. 
We will not follow the way of Axel Olrik and Celander in 
our investigation, who treat the modern traditions as equi¬ 
valent and even superior to the documents of the heathen 
period, because we think it methodically objectionable to 
explain the religious representations of one period by those 
of another. This holds still more true in the present case, 
where the gulf of the conversion to Christianity lies between 
both periods. The only possible way seems to me to scrutinize 
the extant Old-Norse material and if the conclusions, we 
arrive at, are in any way concordant with the modern 
popular tradition, we may consider it as a nice confirmation 
of our deductions. If, on the contrary, our results do not 
agree with it, we must not try to mend them by the use of 
the popular material, but we must seek for the solution of the 
quite independant problem of the origin of the modern folk¬ 
lore traditions about Loki. 



1. The myth of the creation of mankind 

Three gods, vis Othin, Hoenir and LoSurr, find according 
to st. 17 of the Voluspa, two pieces of wood, called Askr and 
Embla, from which they create two living beings, surely a 
man and a woman. Each of these gods has a special gift 
to bestow upon them. The following stanza tells us: 

Qnd l au ne atto 
la n6 lseti 
ond gaf 6 Sinn 
la gaf LoSurr 

65 Jau ne hofSo, 
ne lito go5a; 

65 gaf Hoenir, 
ok lito go5a. 

Unfortunately the meaning of some of the terms, used in 
this stanza, is not sufficiently clear, and several translations 
have been proposed. Qnd of course means »breath», and odr 
»mental activity* 1 ; litir godar may be rendered by »good, 
healthy colour*. But la, and Iceti are obscure. If we compare 
with the stanza the tale, Snorri gives in ch. 8 of G-ylfaginning, 
we find on the one hand a close resemblance, on the other 
remarkable differences. 2 Here we learn about three gods, 
that are called now »the sons of Burr*; then the tale runs on 
in this way: gaf enn fyrsti ond ok Ilf, annarr vit ok hrcering, 
HI dsjonu, mal ok heyrn ok sjon. If the two myths are 

1 Cf. my paper Contributions to the Study of Othin etc. (FF Com¬ 
munications Nr. 94) p. 30 sqq. 

2 Cf. Miss Gras 1. c. p. 4 sqq. 

FFC 110 

Chapter IV. Loki and Othin 


fundamentally the same — and it is after all in a few minor 
details, that they differ from each other — then the gifts 
of the gods may he compared, and in this case the last god 
has given the outward appearance and the senses (speech, 
hearing and sight). With litir godar the word asjonu is in 
concordance; if we translate Iceti by »voice» 1 , we find the 
same idea as in the word moil of Snorri’s tale. Then the 
remaining term la, for which several translations have been 
proposed 2 , may be taken in the sense of »blood» or of »vital 
warmth», both meanings being after all equally problematic. 

In truth, this stanza of the Voluspa reveals to us the 
existence of a creation-myth, but in a form so very obscure 
and incomplete that we can not arrive at a clear insight 
into its structure. I will draw the attention to some difficul¬ 
ties. Othin gives the breath and Hoenir the mental activity. 
That Othin as god of death also bestows the principle of 
life, is, indeed, not at all contrary to what we might expect to 
find, but the combination of Hoenir with the idea of mental 
activity is highly surprising. In fact, what we know about 
this god — and it is to be sure exceedingly little — tends to 
show that his capacities, if he had any, did not lie in the 
intellectual field. We have only to remember the curious 
myth about his being sent as a pledge to the Vanir, where 
his stupidity was easily discovered 3 , to consider it rather 
unpromising for the newly created human beings, that 
such a god should have given to man his mental faculties. 
On the other hand one might expect that 6dr, a word that 
I in my above mentioned paper suggested translating as 
»mental faculties of a higher order, such as poetic genius, 
ecstasy», should have been the special gift of Othin, the 
more so as the name of this god and the word odr are closely 

1 Cf. Finnur Jonsson , Lexicon Poeticum 2 p. 388; S. Nordal, 
Voluspa gefin ut me6 skyringum (Reykjavik 1923), p. 53. 

2 Sijmons-Gering , Edda-Kommentar I, p. 21. 

3 Ynglingasagach. 4. 


Jan de Vries, The problem of Loki 

FFC 110 

connected. Can we therefore assume, that the necessity 
of composing a line with three alliterative words, has induced 
our poet to use the words odr and ond as well as the name 
of Othin in one and the same line and that he consequently 
to avoid the clashing of the cognated words odr and Odinn, 
has made a shift by combining Othin’s name with the word 
ond and by ascribing the odr to Hoenir? 

This is not the only difficulty. The gods find two trees 
and they wish to transform them into human beings. We 
would expect them to begin by bestowing the outward 
appearance, then the principle of life by which these objects 
become animated beings and lastly the intellectual activities, 
by which man is distinguished from the animals. Othin 
and Hoenir have bestowed the last two gifts; the third god 
consequently must fill the remaining gap; it is Lo&urr who 
gives the human form. Here again the order, in which the 
gods bestow their gifts, is different from what we might have 
expected; the poet mentions in the last place what actually 
comes first. Moreover Loburr gives more than we want him 
to: the litir godar may be a circumlocution for the human 
body, but on the other hand Iceti, if it means »voice» and la, 
if it denotes »blood» or »warmth», belong properly to the 
domain of Hoenir and Othin. Perhaps the poet had some 
difficulty in finding words alliterating with Lodurr, or may 
be he did not take these things so seriously; at any rate we 
have ample evidence that the wording of this stanza is 
not so very exact that we can rely upon it for a reconstruction 
of the underlying myth. 

Is it a genuine heathen myth after all? Several scholars 

have doubted it. The distinction between ond and odr 


seems somewhat philosophical and may be ascribed to 
foreign influence. Of course Christian examples have been 
adduced to explain this curious myth; so E. H. Meyer com¬ 
pares the gifts of the three gods with the creation myth in 
G-enesis, however by assuming a reinterpretation of medieval 

FFC 110 

Chapter IY, Loki and Othin 


exegetes. It is unnecessary to prove the improbability of 
this assumption. But it may be possible that this Old- 
Norse myth has its roots in the soil of classical speculations. 
In a contribution to the »Studier tillagnade Axel Kock» 
the Swedish scholar Johan Palmer 1 has tried to point out 
the possible source of the creation-myth in an Arabian 
treatise about astrology. In this treatise the birth of a child 
is supposed to depend upon the planetary gods, and accord¬ 
ingly to every one of these gods a special influence is attribu¬ 
ted: Jupiter gives the breath, Mars the blood, the Sun the 
vital warmth a. s. o . 2 After referring to this learned specu¬ 
lation, Palmer continues: »The likeness between this com¬ 
munication and Voluspa is striking and it is certainly not 
by mere chance that it is the most essential part, the creating 
of man from four elements, which is common to both». One 
can not help from rubbing ones eyes on reading this unex¬ 
pected statement. Between seven or nine planetary gods 
and four elements there is, after all, some difference. More¬ 
over in the Eddie myth there are not four elements but three 
gods. Palmer only arrives at the number four by taking the 
»raw material)), represented by Ask and Embla, as denoting 
the earth and its vegetative forces. But this is quite un¬ 
warranted, as we destroy in this way the whole frame of the 
Old-Norse myth. The poet of this stanza of the Voluspa 
has not had any idea in mind of different elements from 
which man could have been composed and he even could 
not have done so, because he relates quite a different creation- 
myth: not the forming of man from the four principal ele¬ 
ments, but the animating of a lifeless object. The specula¬ 
tions about the meaning of Hcenir and Loburr on the base 

1 Till Vqluspa on pp. 108—118. 

2 In the Arabian treatise there are nine different gifts according 
to the nine months of gravidity. But as Jupiter and Saturnus each 
are mentioned twice, this can not be the original form; cf. Palmer 
p. 109. 


Jan de Vries, The problem of Loki 

FFC 110 

of this venturesome hypothesis do not bring us any nearer 
to the solution of the problem. 

Still the feeling of uneasiness about this Old-Norse myth 
remains. If we remember that Marcus Aurelius distinguished 
three different parts in the human being: body, soul and 
mind, and that both philosophers of ancient times and of 
the Middle-Ages have enounced the same or similar ideas, 
we may ask if the northern poet of the 10th or 11th century 
has had the opportunity of reading some medieval treatise 
about these philosophical speculations. It is possible, 
although it will be difficult to point out his source. In his 
profound and stimulating book »God, Man and Epic Poetry» 
H. V. Routh has taken this possibility also into consideration, 
but he hesitates before this conclusion, as we find an idea 
about a multiple soul also among primitive peoples. And 
if we consider the differences between the Eddie psychology 
and f. i. Dante, we may think it most likely, as Routh 
styles it: »that Northern and Southern Europe were aiming 
independently at a philosophical explanation of the same 
mystery, perhaps helping and confirming each other on the 
way, or perhaps reaching the same conclusion by quite 
separate paths)). 1 

This is, me thinks, a sound reasoning. There is only one 
objection to be made. The author is of opinion that a creation 
myth as narrated in the Voluspa is due to a development, 
only possible on a rather high cultural level. But this is 
not necessary, for we find among primitive peoples specula¬ 
tions about the origin of the first man, that in no respects 
yield to those of the Eddie poem or even to the Greek philo¬ 
sophers. Every one who has read the interesting book of 
Paul Rodin »Primitive Man as philosopher* will have the 
examples ready to hand. It may suffice to quote an origin 
myth of a Winnebago clan, where the following is related: 2 

1 Cf. II. p. 43. 

2 Cf. p. 236. 

FFC 110 

Chapter IV, Loki and Othin 


Earthmaker has created the world and wanted to make a 
being in his own likeness. So he took a piece of clay and 
made it like himself. Then he talked to what he had created, 
but it did not answer. He looked at it and saw that it had 
no mind or thought. So he made a mind for it. Again he 
talked to it but it did not answer. So he looked at it again 
and saw that it had no tongue. Then he made it a tongue. 
Then he spoke to it but still it did not answer. He looked at 
it and saw that it had no soul. So he made it a soul. Then he 
talked to it again and it very nearly said something, but it 
could not make itself intelligible. So Earthmaker breathed 
into the mouth and then talked to it and it answered. 

As Radin subsequently points out, this is a fine example 
of the efforts to explain the enigma of man’s existence in a 
reasonable way. First the newly created man is endowed 
with thought, then with the mechanism for speech, with 
the soul and finally with intelligence. This primitive philo¬ 
sopher obviously has been puzzled by the difference between 
man and animal; a being may have the faculty of speech 
and still not be human; it is the last endowment of intelli¬ 
gence which makes really man. 

So I see no objection to assuming the myth in the Vo- 
luspa to be exclusively Teutonic and purely heathen. Even 
in its sophisticated form it is not incompatible with the 
reasonings of primitive mind. Still I can not subdue my 
doubts about its originality. The creation of man from a 
tree or a piece of wood is a belief of world-wide existence. 
According to Hesiod Zeus made the human species from 
an ash-tree 1 , which curiously accords with the Old-Norse 
poem, where the male tree also is called Askr. In Austra¬ 
lian myths this tale is very frequent 2 : so in the Banks Islands 

1 Works and Days 11.143—145. 

2 See many instances in Dixon, Oceanic Mythology, p. 30 (Island 
of Nieue), 106 (Melanesia), 110 (New Britain and Solomon-islands), 
168 (Eastern islands of Indonesia). 



Jan de Vries, The problem of Loki 

FFC 110 

it is told that Qat cut wood of dracaena trees into human 
shapes; he formed legs, arms, trunks, heads and added ears 
and eyes; then he fitted part to part and finally by means of 
magic practices gave them life. 1 In the same way the 
American tribe of the Qu’iche tell that men are created 
from wooden dolls, who could move and even procreate, 
but they had neither heart nor intellect and roamed about 
like animals. 2 In the Eddie myth we meet however the 
curious representation that three gods make men from 
wooden logs by endowing them each with different qualities, 
roughly speaking, with a human form, with the principle 
of life and with intellect. 

The Eddie creation-myth has been compared by F. von 
der Leyen 3 to a curious folk-tale, in which three or four 
persons shape a woman from a log of wood. In the oldest 
form of this tale, the Vikramacarita 4 , a woodcarver, a 
goldsmith, a weaver and a priest travel together. Once 
they pass the night in the open air and they keep watch 
each in his turn. The woodcarver takes a piece of sandalwood 
and carves it into the image of a beautiful girl. Then the 
weaver dresses her, the goldsmith adorns her with all kind 
of jewels. Finally the priest speaks a magic formula by 
which it is transformed into a living being. According to 
the Indian predilection for casuistic tales, the question now 
arises: whose wife ought she to be? The tradition of this 
tale has many ramifications; it has travelled eastwards to 
Further India and Indonesia 5 , but also to the peoples living 

1 Codrington, The Melanesians p. 157. 

2 W. Iirickeberg, Marchen der Azteken und Inkaperuaner p. 125. 

8 Das Marchen p. 123. 

4 Cf. J. Hertel, Indische Marchen p. 178 sqq. This Indian work 
itself has not come down to us, but it has been incorporated into 
Hemavijaya’s Katharatnakara (written in 1600 and 1601), and 
translated into Mongol under the name of Ardshi Bordshi. 

5 Cf. Bolte-Polivka III, p. 53 sqq and J. de Vries, Volksverhalen 
uit Oost-Indie II, Nr. 184 and note on p. 396. 

FFC 110 

Chapter IV, Loki and Othin 


west of India, as is testified by different Persian, Arabian 
and Turkish versions. 1 It found its way also into Europe, 
although it is found here only very sporadically 2 , obviously 
by way of literary tradition. 

Although those forms of the tale, where only three persons 
cooperate in the making of the girl, show a remarkable 
likeness to the Eddie poem, it seems to me impossible, that 
there exists any connection between the two traditions. The 
Indian tale, a curious example of the adaptation of popular 
tradition to a highly sophisticated literature, can not have 
reached Iceland as early as the 10th century;,on the other 
hand does the Old-Norse myth not go beyond what we 
might expect even in more primitive societies than the 
Icelandic was at that time. Such tales may be invented 
independently in different regions of the world. But the 
comparison with the Indian tale may help us in formulating 
more exactly the peculiar character of the Old-Norse myth. 
The poet of the two stanzas in the Voluspa does not care 
much about the material part of man’s creation; his main 
interest is for the psychical constitution of the human kind. 
We might ask if these stanzas reflect a real myth. If so, 
it must originally have had another form, or else the poet 
has only chosen those details he thought most appropriate 
for his design. In any case it seems to me that we can not 
consider these stanzas as a mere poetical fancy, because we 
can not understand how the poet hit upon such uncommon 
names as Hoenir and Loburr. When Snorri gives a rendering 

1 Cf. Bolte-Polivka l.c. and R. Basset , Mille et un Contes, Recits 
et Legendes Arabes II, Nr. 65, p. 312—313. An African variant is 
printed in L. Frobenius, Atlantis III, Volksmarchen der Kabylen 

p. 100—101. 

2 There are a few examples from Eastern-Europe (cf. Bolte- 
Polivka 1. c. p. 56 and A. Solymossy , Hongaarsche Sagen, Sprookjes 
en Legenden Nr. 69) and a very short German variant (' Wisser , 
Plattdeutsche Volksmarchen p. 99U 


Jan de Vries, The problem of Loki 

FFC 110 

of the myth in his mythological treatise, he — remarkably 
enough — denotes the creating gods as the three, sons of 
Burr. He, however, knew the Voluspa very well, as he 
quotes it on several occasions. Has he known perhaps a 
text of this poem, where these stanzas were wanting and has 
he got his creation-myth from an other literary source or 
from oral tradition? This is not probable, as the likeness 
between both variants is so great as to presuppose a mutual 
dependence. Has he then changed the names on purpose, 
because Hcenir and L65urr were incomprehensible names 
for the public of his days? But a mythographer, such as he, 
might have been glad to add some more details to his picture 
of old heathen belief. 

As a matter of fact, Snorri nowhere mentions the name 
of LoQurr. We find this god, besides in this stanza of the 
Voluspa, also in the Islendingadrapa of Haukr Valdisarson 1 
where a poem is called lid Lodurs vinar. If this poet, who 
lived in the 12th century, has known the peculiarity about 
Loburr that he was the friend of Othin, he can not have got 
this knowledge from the Voluspa, where nothing of this 
kind is said. Perhaps it may be a combination of the three 
gods with the sons of Burr mentioned by Snorri. Still it 
is possible that there have been more traditions current 
about Loburr, and we are once more reminded of the easily 
forgotten fact, that what we possess of Old-Norse literature 
— although it is in itself considerable enough — is only a 
small part of what has been once in existence. In trying to 
»explain» this enigmatical god, we must be very careful, 
as we practically know nothing about him. 

This one fact seems undeniable, that he plays a r61e in 
the Old-Norse creation-myth. It is unthinkable that a poet 
has composed these curious stanzas, only led by the desire 
to relate a quaint and mysterious sounding myth. And 

1 F. Jonsson, Den norsk-islandske Skjaldedigtning I B, p. 539. 

FFC 110 

Chapter IV, Loki and Othin 


finally, against the often repeated assertion, that Lobarr 
is nothing else but a camouflaged Loki, we must insist upon 
the clear wording of the poem, where not the slightest 
allusion to a possible identity of Lo&urr and Loki is to be 
found. How indeed to explain, that a poet treating of Loki 
as one of the principal enemies of the gods in Ragnarok 
should have identified with him a god like Loburr, who 
together with Othin and Hcenir in the beginning of the 
world had created mankind? 

2. The myth of |>j a z i 

The myth of ]>jazi is told in the so-called Haustlong, a 
poem composed by kjobolfr or Hvini at the end of the 9th 
century. It is difficult to understand in detail, but in the 
whole there is no uncertainty about the subject-matter. In 
the Snorra-Edda the same myth is given at fuller length 
partly based on the poem \ partly drawn apparently from 
slightly different sources. For the coarse ending, where 
Skabi’s wrath is appeased by Loki’s obscene performances, 
there is no parallel whatever in the Haustlong. The tale 
about ttjazi consists of a series of popular motives well 
known in European folk-tales 1 2 and it seems difficult to 
find the real myth underlying this work of fancy. As Miss 
Gras has pointed out, there seems to be, in Haustlong as 
well as in the Snorra Edda, a contradiction between the 
important and even favourable part Loki plays in this tale 
and his usual conception as a wicked foe of the gods. She 
thinks it possible that the poem still preserves some recol¬ 
lection of an earlier time when Loki had quite a different 
character and was more intimately connected with Othin. 3 

1 Miss Gras , 1. c. p. 15—16. 

2 Cf. V d. Leyen p. 32 sq and Gras p. 18. 

3 L. c. p. 17. 


Jan de Vries, The problem of Loki 

FFC 110 

If the Skaldic kennings, as Axel Olrik has asserted 1 
possess more value as mythological evidence than the more 
individualistic poetry of the Edda, then circumlocutions as 
Hrafnasar vinr (st. 4), Hcenis vim (st. 7) and pors ofruni 
(st. 8) are of great importance and strangely contradict the 
remark about Loki in st. 12: sldcm sveik opt psu leikum. 
Moreover the only real hero of the tale is Loki, the two other 
gods doing nothing at all. The reconstruction of the original 
form of the myth — if there ever has been any myth at all — 
becomes very difficult, if we 1. must consider the subject- 
matter itself as the result of a later development under the 
influence of popular tales, 2. must reduce the three coope¬ 
rating gods to one actor. 2 The difficulty becomes still greater, 
if we confront this tale with the very rare cases, where 
fcjazi is mentioned in Old-Norse literature. Kormakr 
Qgmundarson says in his Sigurdardrapa, written about 
960: veltu god Pjatsa (st. 6); this expression seems rather 
inappropriate for the only incident in the tale to which it 
can refer: fjazi in his persecution of Loki is caught by the 
gods who make a fire, in which he burns his wings. This is 
more a case of self-defence, than of tricking; it seems to me 
that the wording of Kormakr is more fitted for a tale where 
the gods are the assailing party and have got the better of 
bjazi by means of a ruse. The death of the giant is, at any 
rate, not the most interesting part of the myth; it is therefore 
the more surprising that it has been extolled as a famous 
deed of Thor. For this god says himself in st. 19 of the 

Ek drap }>iaza, inn Jmidmodga iotun 

upp ek varp augom Alvalda sonar 

a t>ann inn heida himin. 

1 Neue Jahrbiicher fur das klassische Altertum, Geschichte und 
deutsche Literatur und fur Padagogik 1918, p. 38—48. 

2 The opinion of Miss Gras on p. 18 that the names of the three 

FFG 110 

Chapter IV, Loki and Othin 


This is again in contradistinction to the Lokasenna, where 
Loki says to SkaSi wishing to hurt her by the recollection 
of her father’s death (st. 50) 

fyrstr ok ofstr var ek at fiorlagi 

fiars ver a f>iaza firifom. 

Moreover this does not agree with the form of the myth in 
the Snorra Edda, where it is Othin, who throws the giant’s 
eyes to heaven, where they become brilliant stars. That 
there have been more tales current about fjazi is not 
impossible; perhaps the Hervararsaga has preserved a faint 
trace of such an unknown myth, where it is related, that 
Sigrlami fell in battle agains Jijazi and that his son Svafrlami 
in his turn slew the giant and took his daughter Fri5r as his 

It may be granted that this saga does not seem very 
reliable in this case, as the giant can not have been killed 
by the gods and by Svafrlami at the same time. 1 At all 
events we have to allow for the possibility that there have 
been several myths about f*jazi and that only one of them 
has come down to us, unfortunately in a distorted form 
that shows the all pervading influence of a popular tale. One 
thing, however, seems to be clear: the principal adversaries 
of the giant are either Loki or Thor, but with the exception 
of the Snorra Edda and Haustlong no other source reveals 
anything about the participation of Othin or of Hoenir. 
How can this be explained? 

In the first place we must pay attention to the singular 
fact that the »myth» of I>jazi is wholly composed of several 

gods belong to an old myth, is not in accordance with her own remark, 
that Loki is the only actor in this tale. This assertion remains without 
any convincing proof. 

1 Can there be any relation between this Svarflami and the 
Othin-name Svafnir? It seems to me highly improbable and such a 
hypothesis can only make the problem still more complicated. 


Jan de Vries, The problem of Loki 

FFC 110 

folklore-motives which occur in other mythological tales as 
well. After the episode of the giant-eagle that carries away 
the best morsels of the gods’ dinner, we meet the motive of 
the miraculous object to which every one sticks. It is found 
in a slightly different form in the tale of GeirroSr; according 
to the Snorra Edda Loki in the valshamr of Frigg flies to the 
hall of the giant, but can not free his feet from the window 
where he has alighted. Just as in the myth of I>jazi he must 
promise to put I&unn in possession of the jotunn, in the tale 
of G-eirrobr he has the still more difficult task of bringing 
Thor into the giant’s hall without his dreaded hammer 
and his magic belt. 1 After having performed this mischief 
Loki is compelled by the gods to repair the fault; this is 
the same motive as occurs in the myth of the giant who had 
boasted of being able to build a palace for the gods. Loki 
flying by means of the valshamr of Frigg or Freyja is a 
Commonplace incident of Scandinavian mythology; we 
found it already in the tales of J>jazi and G-eirroSr and we 
may add in the frymskviba. These last mentioned myths 
belong to the cycle of Thor. Finally the motive of two 
supernatural beings pursuing each other in the disguise of 
birds, the god making a narrow escape just inside the wall 
of Asgard, is told with different details in the myth of Othin 
who steals the poetic mead from Suttungr. 

So I think it probable that the tale of jjjazi is composed 
by borrowing typical motives from other myths; these 
belong as well to the person of Othin as to that of Thor. 
This makes it sufficiently clear that we should not waste 
our time in trying to make a reconstruction of the original 
form of the myth about l>jazi. The only part that may be 
authentic, as we do not find any parallels in the mythological 
literature of the Scandinavian peoples, is the disappearance 
of Ibunn and her apples of life; although a possible influence 

1 Snorra Edda I p. 284 sqq. 

FFC 110 

Chapter IV, Loki and Othin 


of classical tales can not be altogether excluded, we may 
suppose this to be one of the characteristic features of the 
hjazimyth. Perhaps we may go one step farther: the fruit 
of life belongs naturally to the sphere of Othin, as he is 
intimately connected with the mythical mead, a drink that 
bestows not only the gift of poetry, but even of immortality. 1 
So we may arrive at the conclusion, that Othin and Loki 
probably belong together in this tale, although the testimony 
of the HarbarbsljoS again points to the possibility, that in 
another form of the myth Thor is responsible for the death 
of the giant. As to Hoenir, there is no place for him in this 
myth; his only task is to make complete the sacred number 
of three. He appears also in a few other tales together with 
Othin and Loki, to which we will now turn our attention. 

3. The introduction of Reginsmal. 

In the prose-tale, preceding the Reginsmal in the poetic 
Edda, the same three gods get into difficulties by their 
having caught and slain an otter. The fact that the intro¬ 
duction of this poem has been added in Scandinavian tradi¬ 
tion to the heroic legend of the Nibelungs does not exclude 
its once having been an originally independent mythical 
tale. This seems, at any rate, to be the opinion of Miss Gras , 
who even tries to arrive at some idea about its primitive 
form. 2 

This question, however, can not be settled in such a 
summary way. Is there any indication of its originality? 
The form, in which it is found in Reginsmal, shows clear 
signs of an adaptation. If we ask, what might have been the 
oldest conception: the dwarf Andvari or the Andvaranautr, 
the answer seems to be not so very difficult. The ring, which 

1 Cf. my De Germaansche Oudheid, ch. XLI. 

2 L. c. p. 14. 


Jan de Vries, The problem of Loki 

FFC 110 

has been from the very beginning the tragical object in the 
lays of Sigurd, was supposed to be laden with a curse and 
quite in harmony with the later tendency to lengthen a 
heroic legend backwards as well as forwards, a tale was 
subsequently invented to explain .this singular curse. This 
ring was really an andvaranautr, that is literally: »a precious 
object which causes terror or grief». It is only natural that 
in course of time such a name was interpreted as a gift 
of a mythical being with the name of Andvari. Then a tale 
had to be found or imagined to account for the origin of the 
tragic gift. If we look at the Eddie poem itself, we see that 
it is partly composed of motives, appearing elsewhere as 
well. The most curious instance is the fishing of Loki, 
who wants to catch a pike and borrows a net from the 
seagoddess; we are singularly reminded of an experience he 
had himself, when he was caught in the form of a salmon 
by Thor. If there ever has been a myth about the super¬ 
natural origin of a precious ring, then its contents may have 
been quite different. Perhaps we find a trace of it in the 5th 
stanza of Reginsmal, where the dwarf Andvari calls a certain 
otherwise unknown G-ustr the real owner of the ring. We do 
not know if this person has been in possession of the precious 
object before it came into the hands of Andvari, as Gering 
suggests \ or if Gustr and Andvari are only different names 
for the same mythical being, as F. Jonsson is inclined to 
suppose 1 2 ; perhaps Sijmons is right in assuming that this 
curious stanza belongs to another poem treating the same 
legend. When he furthermore asks, if the eight heroes 
mentioned in the following lines have anything in common 
with the Sigurd-legend proper, he has once more called 
our attention to the confused and unreliable form of the 
traditions about the, Andvaranautr. 

1 Edda-Kommentar II, p. 167. 

2 Lexicon poeticum p. 209 »sikkert identisfo). 

FFG 110 

Chapter IV, Loki and Othin 


Is is a well-known fact, that the myth about the covering 
up of the otter’s hide with gold presents several difficulties; 
we may compare it with old German customs of atonement 
for the slaying of an animal as has been amply illustrated 
by Jacob Grimm. 1 But there are again so many differences 
that it is unwarrantable to explain the curious tale of 
Andvari by assuming that the same custom has been known 
in Scandinavia. 2 The resemblance to the tale of Fredegar, 
that the Goths had to pay a penalty to the Franks consisting 
in heaping up so much gold until it reached the top of the 
head of a Frankish warrior sitting on his horse, is extremely 
great. There is even this detail, that the receiving party 
is not content, until the gold reaches as high as the summit 
of the warrior’s lance, just as HreiQmarr wants one hair 
that is still bare, to be covered up with the gold. I do not 
think it probable that there is any immediate connection 
between the Frankish tale and the Old-Norse poem, as 
G. Schiitte seems to believe 3 ; but in my opinion it is clear 
that we have here a literary motive, that consequently may 
have been added to the Old-Norse tradition of the Nibelungs 
in a rather late period. 4 It is surely of no small importance 
that there is no allusion to this myth in skaldic poetry before 
Snorri coins the kenning otrgjold in a stanza of his highly 
artificial Hattatal, where he uses not less than four different 

1 Deutsche Rechtsaltertumer II, p. 239 sqq. 

2 A. Raszmann in Germania XXVI (1881), p. 378 sq. 

3 Cf his paper in Edda, Nordisk Tidsskrift for Litteraturforskning 
1917 II p. 249—250. 

4 As it is unknown in the German tradition of the Nibelungs, we 
have no reason to suppose that Scandinavia acquired from Germany 
a form of the Nibelung lay with this motive. But it is quite possible 
that both traditions wandered to the North, independently and at 
different periods; then they came into contact with each other 
because they were both of Frankish origin and hence in the mind of 
Old-Norse people belonged together. 


Jan de Vries, The problem of Loki 

FFC 110 

circumlocutions for gold. 1 Such a list of kennings is found 
also in the Bjarkamal 2 and it is even probable that Snorri 
has imitated some of them, cf. his reidmalmr Gnitaheidar with 
Rlnar raudmalmr in Bjarkamal, or ftungfarmr Gratia with 
Grana fagrbyrdi. Here we find also the kenning dpi tregum 
Otrs gjoldum, but this does not prove that the myth is old, 
for these stanzas do not belong to the oldest form of the 
poem but to the skaldic art of the 12th century. 3 The next 
and at the same time the last allusion is to be found in a 14th 
century poem of Einarr G-ilsson 4 , who uses the same kenning 
as Snorri,but onlyin amore complicated form: Otrs naudgj old. 
These sporadic records in the skaldic literature do not prove 
at all that the myth itself is old and genuine heathenism. 5 
Moreover the fact that the tale as it appears in Reginsmal 
is only related as a prosaic introduction and that it is filled 
up with stanzas belonging to a gnomic poem like Alvissmal 
or VaffiruSnismal make it still more credible that it is a 
relatively recent accretion to the legend of the Nibelungs. 

Then the divine triad appear in a very unexpected light. 
Othin belongs as the god of war to a hero like Sigurd; that 
is in conformity with Old-Norse conceptions. Loki as the 
cunning god may have been chosen for the part of the person 
who gives good advice in a difficult position. We will see 
presently that he often does so. But what has Hoenir to do 
with this heroic matter? There is indeed no reason whatever 
to presume that he should have had more importance in an 

1 Stanza 41, cf. F. Jonsson, Skjaldedigtning II B, p. 72: besides 
otrgjold three other circumlocutions. 

2 F. Jonsson, ibid. I B, p. 170, st. 4—6. 

3 Heusler-Raniseh, Eddica Minora p. XXV and Axel Olrik, 
Danmarks Heltedigtning I, p. 99. 

4 Skjaldedigtning II B, p. 423, st. 19. 

5 E. A. Kock reads in st. 10 of the Porsdrapa the kenning arfi 
eirs fjardar for Loki and explains it as an allusion to the Andvaranautr 
(cf. Notationes Norroenae § 455). But the ms. has arfi eids fjardar 
(cf. p. 60). 

FFC 110 

Chapter IV, Loki and Othin 


original form of this quasi-myth 1 ; this »deus otiosus» has 
only to give his name in order to make the triad complete. 
And I think that we are allowed to put the question, why 
the poet hit upon the name of this quite unimportant god. 
The only explanation seems to be that he has read a tale 
somewhere, where Hcenir is mentioned together with Othin 
and Loki. But the only tale where this is the case, is the 
myth of f>jazi. Here Hoenir appears in exactly the same 
circumstances as in the introduction of the Reginsmal. It 
would indeed be rather miraculous if two quite independent 
and original myths had been known, in which Loki was 
the principal actor, Othin a subordinate assistant and Hcenir 
a mere name. So it seems to me not borne out by the extant 
material that the triad of gods, as they appear in a couple 
of literary traditions, answers to a corresponding mythical 

4. The three gods in popular tradition. 

A Faroese ballad, where Othin, Hirnir and Loki are 
mentioned together, has had a considerable importance 
in the speculations about the meaning of Loki. In the Lokka 
Tattur 2 a peasant must try to save his son from the pursuit 
of a giant; he asks the help of the three gods to hide him from 
the troll. Othin puts him into a grain of wheat, Honir 
changes him into the feather of a swan, but the giant finds 
him in both cases. Then Lokki changes him into an egg of 
the spawn of a fish. Although the giant even finds him here, 
the egg falls out of his hand; Lokki seizes it and succeeds by 
means of a trick in killing the monster. 

Again we are in the presence of a typical folkloristic 
motive and Axel Olrik has already made the observation 

1 Gf. Grass 1. c. p. 14. 

2 V. U. Hammershaimb , SjurSar kvae&i p. 140—145. 


Jan de Vries, The problem of Loki 

FFC 110 

that the mythological value of this ballad is practically nil 1 . 
The fact that the gods act on behalf of a peasant’s son, is 
sufficient to show that this humoristic tale has nothing to do 
with heathen mythology, but belongs exclusively to the do¬ 
main of popular literature. It is characteristic of Paroese 
traditions, where the names of the gods have been preserved 
in folk’s memory down to relatively recent times, that the 
poet has tried to make the tale interesting by introducing 
three gods into it. We must not expect, however, to learn 
anything about the character of Loki, for, as Olrik remarks, 
he is the same as we find him everywhere in mythical tales: 
benevolent to mankind but wily in his relation to demons 2 . 

There remains only one question to be answered: Prom 
which-source did this ballad-poet obtain the names of this 
curious triad? Of course not from any heathen myth, 
unknown to us, but from the common stock of Old-Norse 
literature. If we take into consideration the keen interest 
of Faroese ballad-poets for the legend of the Nibelungs and 
the way in which they have made use of written saga’s, 
we can not arrive at any other conclusion but the supposition 
that he has found the three gods Othin, Hanir and Loki 
in the introductory prose of Reginsmal. For our knowledge 
of the heathen religion and mythology the Lokka tattur is 
absolutely worthless. Those scholars who have made use of 
it in their investigation into the character of Loki have been 
misled by their uncritical confidence in the traditional 
character of popular tales and ballads. 

The same applies to an old-English incantation-formula 
that is brought into connection with this group of mythical 
tales. The clergyman Robt M. Keanley tells about an experi- 

1 Loki i nyere folkeoverlevering, Danske Studier 1908, p. 196 

2 Miss Gras is quite on a wrong track if she considers »this more 
recent poem more original in several respects than the introduction 
to Reginsm&l and the Snorra Edda» (p. 22). 

FFC 110 

Chapter IY, Loki and Othin 


ence which he had, when still a boy, in Lincolnshire, where 
he was born 1 . During an epidemic fever he came with qui¬ 
nine to an old woman whose grandchild was very ill. She, 
however, took him to the bedroom and »there in the centre 
of the footboard were nailed three horseshoes with a ham¬ 
mer fixed crosswise upon them. Taking down the hammer 
she smartly tapped each shoe, saying words to this effect 
as she did so: 

Father, Son and Holy Gfhost 
Nail the devil to this post — 

With this mell I thrice do knock 

One for God, and one for Wod, and one for Lok». 

But perhaps it is better to quote the formula in the dialectic 
form as it has been published afterwards 2 : 

Feyther, Son and Holy Ghoast 
naale the divil to this poast; 

Throice I smoites with Holy Crok 

with this mell Oi throice dew knock 

One for God an’ one for Wod an’ one for Lok. 

Axel Olrik points out the identity of Wod and Lok with 
the Old-Norse Othin and Loki; and pretends that God has 
taken the place of Hoenir. If we then take into considera¬ 
tion the Faroese ballad we may arrive at the conclusion, 
that there existed a magical formula with these three gods 
and that it has been current from Lincolnshire to the Faroes. 
This formula makes it probable that »en trehed hvori Woden 
og »Lok» havde plads, ogsaa er paakaldt af de Angler der 

1 Folklore 1898 p. 186 = Axel Olrik, Danske Studier 1908, 
p. 200. 

2 County Folk-Lore V, p. 125 = F. Ohrt, Trylleord, fremmede og 
danske (Danmarks Folkeminder XXV) p. 82 note. 


Jan de Vries, The problem of Loki 

FFC 110 

vandrede til Britanniem x . Dr. Philippson is of opinion that 
the Anglosaxon belief in Loki is not proved by this formula, 
but that it is far more likely to have been introduced into 
England from the Faroes, for here the three gods occur in 
the Loka tattur 1 2 . A curious conclusion indeed, but after all 
an alternative which Olrik had also taken into consideration: 
»Dog selv om formlen skulde were overfort fra vikinge- 
tidens Nordboer, giver den en hojst interessant berigelse af 
vor viden: yder den eneste direkte paakaldelse af Loke — 
ganske vist kun som en person af Odins-treheden» 3 . 

It is surprising that no one, before enunciating such 
far-reaching conclusions, has asked if this piece of English 
folk-lore is quite reliable. For one may wonder that a boy 
who once has heard a magic formula under rather nerve¬ 
straining circumstances, recollects it about forty years la¬ 
ter without one single alteration. Moreover is it possible 
that a formula has been handed down during a period of 
fourteen ages without considerable changes in form and 
contents, especially if we consider that it must have been 
originally an alliterative poem and that it has been recast 
into the modern form of a rhyme-verse? F. Ohrt has cast 
doubt upon the trustworthiness of this formula 4 and I 
think it also quite unadmissible to make use of this questio¬ 
nable evidence for the reconstruction of the old Teutonic 
belief. God of course belongs to the Christian belief, Wod 
may denote the heathen god, whom people remembered 
even in Christian times as the principal deity of their pagan 
forefathers. But what of Lok? As a divine triad-is obliga¬ 
tory in this kind of formula, a third one had to be added and 
its name had to rhyme with the words crok and knock in 

1 Danske Studier 1908, p. 201. 

2 Germanisches Heidentum bei den Angelsachsen (Leipizg 1929) 
p. 153. 

1 Ibid. p. 202. 

4 Trylleord, fremmede og danske, p. 82—83. 

FFC 110 

Chapter IV, Loki and Othin 


the preceding lines. Moreover, there are so many fanciful 
words and names in charms, used in modern times, that 
we must he very careful in abusing them as reliable docu¬ 
ments from the heathen period. 

5. LoQurr and Loki 

It is indeed a singular fact that we find the gods Othin 
and Hcenir acting together with a third deity, who is called 
at one time Lo5urr, Loki at another. The conclusion lies 
ready to hand: Loki must be the same god as LoQurr. This 
has been asserted by several scholars, who argued that in 
these mythical tales the same triad was meant and that 
consequently LoQurr and Loki were different names for one 
single divinity 1 . Only very few have expressed their doubt 
about this identification, as f.i. J. Hoffory 2 , F. Jonsson 3 
and J. Palmer 4 . 

Is there any possible means of arriving at a satisfactory 
conclusion? We must begin to dismiss all etymological 
speculations, for they can only confuse the problem. The 
question evidently must be this: are we entitled on the basis 
of the myths known to us, to conclude the identity of Loki 
and LoQurr, although in the whole extent of Old-Norse tradi¬ 
tion there is not the slightest allusion to it? This is the more 
surprising as it is expressly stated on the contrary, that Loki 
is the same deity as Loptr. The Icelandic mythographers 

1 Cf. E. Wisen, Odin och Loke, p. 70; A. Noreen, Tidskrift for 

Philologi NR IV, p. 28 sqq; P- Herrmann, Nordische Mythologie 
p. 405; Golther, Handbuch der germanischen Mythologie p. 409; 
A. Olrik in Festskrift Feilberg p. 587; E. Mogk in Hoops’ Real¬ 
lexikon III, col. 164 b; F. R. Schroder, Geraianentum und Hellenis- 
mus p. 116; Gering-Sijmons, Edda-Kommentar I, p. 22—23; Miss 
Gras 1. c. p. 6 sq. 

3 Edda-Studien p. 117. 

3 GodafraeSi Norfimanna og Islendinga p. 84. 

4 Festskrift/4a:ef Kockp. 113. 



Jan de Vries, The problem of Loki 

FFC 110 

seem not to have been impressed by the strange resemblance 
between the creation-myth on the one side and the diffe¬ 
rent tales about Othin, Loki and Hoenir on the other: Now 
I have tried to make it clear that both the Faroese ballad 
and the English magic formula are absolutely unreliable 
material and that the introduction of Reginsmal in the 
form wherein it is preserved is no original myth at all. So 
there remains only the myth of t>jazi, which presents again 
the serious inconvenience that it is quite submerged by 
typical folklore-motives. We are unable to see, what may 
have been the original form of the tale, when it was still a 
myth. As Hoenir has nothing to do whatever, we do not 
know, if he belonged from the very outset to the deities 
Othin and Loki, or if he has been added to complete the 
triad. In this case the reason of his being chosen may have 
been, that he was found as the third deity in another triad, 
where besides Othin LoSurr was mentioned, whose name 
had some resemblance to Loki. This does not imply, of 
course, that Lo8urr and Loki are the same god. 

The startling conformity between the triads Othin- 
Hoenir-LoQurr and Othin-Hoenir-Loki is in itself not suffi¬ 
cient to prove this identity. The more so, as these myths 
belong to quite different spheres of religious representations. 
Now more proofs have been adduced in favour of the identi¬ 
fication of Loki and L65urr, partly on an etymological, 
partly on a folkloristic basis. As Lo8urr and Loki were 
supposed to be the same deity, scholars tried to explain the 
name L68urr as denoting a fire-demon. 

We possess not less than four etymologies. 1. The word 
may be compared with the German verb lodern 1 and means 
»the blazing one»; 2. The name is derived from an older from 
*V16I)urr and is the same as the Indian Vrtra, the demon of 
heat 2 (this etymology as well as the former one has not been 

1 Wisin, Odin och Loke p. 70. 

2 A. Noreen, Tidskrift for Filologi NR IV, p. 28 and Urgermani- 

FFC 110 

Chapter IV, Loki and Othin 


accepted). 3. The word is explained from a presumable 
name *LuhtmraR l , which should have had the meaning of 
»bringer of flames». 4. An original form Loga]>ore, which is 
found on the runic inscription of the large brooch of Norden- 
dorf, lies at the bottom of the name 2 ; it must be nearly 
related to the word for »fire», logi. Of course these etymolo¬ 
gies have only a hypothetical value and hold good only 
as long as we accept the interpretation of LoSurr as a fire- 
god 3 . 

This may, perhaps, not be said of the last explanation 
and Miss Gras rightly stresses the importance of the runic 
inscription, which, if correctly interpreted, may yield very 
important evidence in favour of the existence of this deity, 
not only in Old-Scandinavian belief, but also among other 
Germanic tribes. Therefore she criticizes Gering, who with¬ 
out any counterevidence rejects'the etymology of v.d. 
Leyen as a mere soap-bubble 4 . But it is indeed very diffi¬ 
cult to refute this, as the assertion itself is not supported by 
any argument at all! For it can not be proved that the word 
logajiore is the name of a god 5 , nor is the etymology of Lo5urr 

sche Lautlehre p. 102 (not repeated however in his Altnordische 

1 This is a suggestion of Max Blankensteiner first mentioned 
by Olrik’s study on Loki in the Feilberg-Festskrift p. 587 and after¬ 
wards adopted by Mogk and Sijmons. 

2 First F. v. d. Leyen , Zeitschr. des Vereins fur Volksk. XXV(1915) 
p. 136 sqq and slightly improved by Von Unwerth, ibidem XXVI 
(1916) p. 81 sqq and W. Krause , Zeitschr. f. d. Alt. LXIV (1927) 
p. 269 sqq. 

3 The etymology of G. Wilke , Die Religion der Indogermanen 
in archaologischer Beleuchtung p. 120, who explains the word as 
Huk-turo-s or »basket of light» is quite unacceptable. 

4 Edda-Kommentar I, p. 23 note. 

5 There are several other ’translations’ of this enigmatic word. 
V . d . L'yen, l.c. renders loga pore Wodan by »Wodan moge die hei- 
lige Flamme entfachen»; P. Herrmann in Altdeutsche Kultgebrauche 
(Jena 1920) p. 65 by »Die Heirat ersiege, Wodan». Quite different 


Jan de Vries, The problem of Loki 

FFC 110 

from a presupposed Hohafioraz more than a Jiypothesis. 
As the Scandinavian and West-Germanic words do not fully 
agree, and moreover as there is no argument at all for the 
identity of these names on the basis of their religious mea¬ 
ning, the ingenious interpretation of v.d. Leyen has to be 
accepted or rejected according to the inclination of each 
scholar. Miss Gras is of opinion that the name of the Dutch 
demon Lodder may be of some value in settling this question; 
I shall presently have the opportunity to show that this is a 
pure mirage. 

In the creation-myth the enigmatic god L65urr bestows 
some qualities belonging to the physical status of mankind. 
If from the very beginning three gods have been included in 
this myth, we might expect that a deity, intimately connec¬ 
ted with procreation and fertility would take part in it. 
A god like Freyr should not be wanting in the creation of 
mankind. If so, this LoSurr, about whom we practically 
know nothing, may fill the place Of Freyr 1 ; could it there¬ 
fore be possible that he is a deity of fertility? With such a 
representation the etymology, proposed by J. Sahlgren 2 
tallies very well; he explains the name by an older form 
*Lo9verr, which he compares witli the name Lodkona to 
be found in the topographical name Locknevi, older Lodh- 
konuwi. This Lo&kona denotes in his opinion »a goddess of 
fertility#; the first syllable lod- may be compared with gothic 
liudan. This etymology has been favourably accepted by 
several scholars 3 

is the reading of S. Feist, Arkiv for Nordisk Filologi XXXV (1919) 
p. 262 sqq, who inverts the order of the letters: ero pa gol, which 
should mean: »Earth sang the charm». 

1 This has already been argued by Defter and Heinzel in Paul 
und Braune’s Beitrage XVIII (1894) p. 560. 

2 Namn och Bygd VI, p. 33 sqq. 

3 F. R. Schroder, Germanentum und Hellenismus p. 116; Giintert, 
Der arische Weltkonig und Heiland p. 309; E. Wesson in Acta Philo- 

FFG 110 

Chapter IV, Loki and Othin 


There is, however, one difficulty; this word must 
have a short vowel o; the actual form in the Edda has a 
long 6. The objection of Gering 1 , that lofjors in the Islend- 
ingadrapa is used in a&alhending with glojta is of no value, 
for Haukr Valdis'arson, who lived in the 12th century has 
borrowed the kenning Lofiurs vinr for Othin from Eyvindr’s 
Haleygjatal, where the vowel may be short as well as long. 
In fact, the Voluspa proves the length of the vowel, as in 
the line of st. 18 lo gaf Lodurr no other quantity is possible. 
But Sahlgren has aptly suggested, that the long 6 may be 
the consequence of the fact, that in course of time the name 
Lo&verr (where the first syllable is by position long) was 
changed into Lo&urr and then the syllable Lod-, used in the 
same line, had to lengthen its vowel. Hence, all things 
considered, the ingenious hypothesis, that explains Lo&urr 
as a god of fertility, leads to very satisfactory results. 

If Lo&urr is such a deity of fertility, his identity with 
Loki, about which the Icelandic tradition has not the slight¬ 
est idea, becomes wellnigh impossible. F. R. Schroder goes 
as far as to contend that on the ground of the »indubitable» 
identity Loki also must have been a fertility demon, but 
I think such a way of reasoning in mythologicis very dan¬ 
gerous. The real character of Loki must become evident 
from his myths, not from hypothetical constructions. 

A fresh support to the theory about the identity of 
Lo&urr and Loki has been adduced by Miss Gras, who takes 
into consideration the Dutch popular tradition about a 
demon called Lodder or Loeke. The idea of their belonging 
to the same religious conceptions as Lo&urr and Loki was 
first expressed by Griiner-Nielsen and Axel Olrik 2 and they 

logica Scand. IV, p. 101; Palmar, Festskrift Axel Kock p. 113; K. F. 
Johansson in Skrift. Hum. Vet. Samfund i Uppsala XX, 1 p. 96 

1 Edda-Kommentar I, p. 23. 

2 Danske Studier 1912 p. 87 —90. 


Jan de Vries, The problem of Loki 

FFG 110 

arrive at the conclusion that the-figure of L65urr-Loki may 
have been known in the Netherlands, but that it can not 
be proved. The possibility depends upon our ideas about 
the original character of these gods. If we explain this mythi¬ 
cal being as »et eller andet natligt gsekkende vsesen, snarest 
af lygtemandsagtig art», we may easily arrive at a defini¬ 
tion, that holds good both for the Dutch and Scandinavian 
mythical beings. But then we ought to add: not for the god 
of Old-Norse tradition, but for the popular demon as it is 
found in modern popular lore. Only by assuming that this 
modern representation is at the same time the original form 
of the Loki-conception — an enormous exaggeration of the 
value of the popular traditions now-a-days — this construc¬ 
tion may be possible. But then we have not paid sufficient 
attention to the following facts: 

1. Loki is nowhere in heathen tradition such an inferior 
demon as the Dutch Loeke-Lodder; his character as a fire- 
god is moreover very questionable. 

2. Lo5urr has nothing whatever to do with fire; he is 
only an actor in the creation myth and is called the friend 
of Othin. 

3. LoQurr can only be compared with the Lodder of 
Dutch folklore if we accept his identity with Loki and in 
consequence attribute to him all the details characteristic of 

4. Only by reducing the real sphere of activity of the 
Dutch Loeke-Lodder on the one hand and of the Scandina¬ 
vian Lokke in popular belief — about whom we will pre¬ 
sently have more to say (ch. XI) — on the other hand to the 
vaguest possible formula: a kind of »arnevsette» or demon of 
the hearth, is it possible to detect any likeness between 
them. But then no trace is left of the heathen god Loki. 

Miss G-ras is of opinion, that the combination with the 
name Logafore on the brooch of Nordendorf is a very con¬ 
clusive argument for Olrik’s construction. A Dutch demon 

FFG 110 

Chapter IV, Loki and Othin 


an Old-German god, the Old-Norse LoSurr, it all seems to 
fit in wonderfully. I see, however, nothing but a medley of 
contradictions. For the Dutch Lodder is a demon of no 
importance at all; his character is explained by Miss Gras in 
connection with the element of water — on quite unsatis¬ 
factory grounds — Lo5urr is a god of some importance, as 
he is the friend of Othin and an actor in the creation-myth. 
This does not tally. Is the Dutch demon perhaps a degenera¬ 
ted form of an original god corresponding to the Eddie 
LoSurr? Or is LoSurr on the contrary a higher developed 
form of an original demoniac being like Lodder-Loeke and 
the modern Scandinavian Lokke? Most scholars will con¬ 
sider the last alternative to be the most probable. But 
then the Old-German (Alamannic?) Loga| ore of the seventh 
century must have been again a highly developed deity, 
as he is mentioned in the inscription together with Wodan 
and Donar 1 , and even takes the first place. What then 
is his connection with a demon like the Dutch Lodder, who 
certainly is no »arnev?ette» at all, but a ghostlike being, 
such as an »alf» or a »kobold»? In fact, the likeness consists 
only in the superficial similarity of the names and when 
these are found in such widely separated geographical and 
chronological areas, we can not be too cautious in basing 
upon such a similarity, which may be only fortuitous, a 
hypothesis about the original character of a god. 

So I will leave this field of barren speculations and re¬ 
turn to the variegated richness of Old-Norse mythology. 

1 If we accept these combinations, the explanation of the enigma¬ 
tic Hcenir is found! Of course he must be the same as Donar, as most 
clearly results from the comparison of the triads Othin-Lo&urr- 
Hoenir and Wodan-Logaf>ore-Wigil>onaR. 


1. The tale of G-eirroQr 

The adventures of Thor on his journey to the giant 
Geirrodr are related in several sources. About the year 
1000 the Icelandic skald Eilifr GoQrunarson treated the 
myth in the borsdrapa, a poem that is rather difficult to 
understand because the poet made use of very complicated 
and obscure kennings and consequently the tradition is in 
many places defective. 1 It has been preserved in the manu¬ 
scripts RWT of the Snorra Edda, immediately after the 
paraphrase of the myth; in U the poem itself is wanting, 
but the tale closes with the remark: »Eptir f>essi sogu hefir 
ort Eilifr Go&rilnarson i f>6rsdrdpa» 2 . As R. G. Boer ob¬ 
serves 3 , this proves that in the archetype of the manu¬ 
scripts the poem has been present; it does not follow how¬ 
ever, that Snorri himself had already added the poem as a 
piece of justification for his tale. 

The myth, as it is told by Snorri, does not fully agree with 
the borsdrdpa. But he had besides this poem another source, 
for he quotes two stanzas from a poem in fornyrSislag. Here 
several names are mentioned, those of the giant’s daughters 
and of the river Thor must cross before arriving at the 

1 F. Jonsson, Skjaldedigtning I A, p. 148—152 and B, p. 139 

2 Snorra Edda II, p. 301. 

3 Aarboger for Nordisk Oldkyndighed og Historie 1924 p. 184. 

FFC 110 Chapter V, Loki as the companion of Thor 


dwelling of GeirroSr. The myth seems to have been well- 
known, for the poet ftodolfr Arnarson makes a visa on the 
demand of the Norwegian king Haraldr har5ra5i (1047— 
1066) in which he makes an allusion to the combat between 
the god and the giant. 1 

In the Gesta Da no rum of Saxo Grammaticus the myth is 
alluded to in the story about the journey of Thorkillus to 
the realm of Geruthus, a kind of Underworld, where the giant 
and his daughters are lying dead after having been over¬ 
powered by Thor. 2 This passage of a couple of sentences is 
of slight value for our knowledge of the myth; the same may 
be said of the romantic saga of forsteinn boejarmagn 3 , 
where the heathen god is replaced by a peasant’s son. 

The German scholar E. Mogk has recently published a 
very valuable paper on these different traditions. 4 He 
arrives at the conclusion that all our sources present the 
myth in a literary form, varying according to the character 
of the authors. As for Snorri, he has combined different 
myths and has composed a tale from this material, which 
Mogk has very aptly called »mythologische Novelle». 5 As 
the two sources of Snorri he considers the i>orsdrapa and 
the poem in fornyrSislag, in which Thor had some adventures 
with female »trolls». But this last supposition seems to me 
improbable as in one of the stanzas (only occurring in U) 
Thor himself says: 

fm er Gjalp ok Gneip doetr GeirraSar 
vildu hefja mik til himins. 

1 F. Jonsson\. c. I A, p. 380 andB, p. 350. 

2 Ed. Holder p. 290. 

3 FmsIII, p. 175—198. 

4 Die Uberlieferungen von Thors Kampf mit dem Riesen Geirrod 
in Festskrift tillagnad Hugo Pipping (Svenska Litteratursallskapet 
i Finland CLXXV, 1924) p. 379—388. 

5 Cf his paper Novellistische Darstellung mythologischer Stoffe 
Snorris und seiner Schule, in FF Communications Nr. 51. 


Jan de Vries, The problem of Loki 

FFC 110 

It is obvious that this is the same story, that we find in 
the f’orsdrapa and so I am of opinion, that the fornyrSislag- 
poem also treated of the myth of G-eirro&r, although in a 
different way from the skaldic poem. The companion of 
Thor is in the Snorra Edda Loki, in the t'orsdrapa however 
J>j41fi; we might ask what it was like in the original form. 

In Eilifr’s poem Loki is mentioned in a few stanzas. This 
is the case in the opening stanza: 

Flugstalla re5 felli 
fjornets go 8a at hvetja 
(drjugr vas Loptr at ljuga) 
logseims fa&ir heiman; 
geSreynir kvad groenar 
Gauts herfrrumu brautir, 
vilgi tryggr, til veggjar 
viggs G-eirrndar liggja. 

The father of the Midgard-serpent is Loki, and he has 
tricked his companion Thor into a very dangerous expedition, 
by saying that green roads conducted the traveller to 
Geirrod’s dwelling. The epitheton vilgi tryggr and the 
parenthetic clause drjugr vas Loptr at ljuga show us Loki 
as a deceiver; this agrees very well with Snorri’s story, 
although it is not necessary to assume that Eilifr has known 
the introductory tale as to how Loki was caught by the giant 
and afterwards released on the condition that he would 
bring Thor unarmed into the house of the giant. 

In the following stanza it is told, that Thor was not 
reluctant to accept Loki’s proposal (I J 6arr let skommum 
GammleiQ biSja sik geSstrangrar gongu 1 ). Moreover it is 

1 According to the interpretation of H. Kuhn, Das Full wort 
of-um im Altwestnordischen p. 37, who proves that the text of 
F. Jonsson: gedstrangr of let gongu is impossible. His reading is more 
satisfactory than that of I. Lindqvist, Norrona Lovkvaden fran 
800- och 900-talen p. 94, who proposes: gedstrangr dr lit gongu. 

FFC 110 Chapter V, Loki as the companion of Thor 


expressly stated that they were both eager to go: fystusk 
t>eir at t>rysta J>orns niSjurn. 

Again Loki is mentioned in st. 3, but here the text 
presents serious difficulties: 

Gorr yard i for fyrri 
farmr meinsvarans arma 
soknar hapts me 5 syipti 
sagna galdrs an Rognir. 

F. Jonsson arrives, although by means of a strange 
rearrangement of words at the following translation: »f jalfe 
was more disposed to go with Thor than Loki was». This 
interpretation is very improbible on the following grounds 

1. In st. 2 it is said, that both Thor and Loki were eager to 
fight with the giants; then st. 3 can not tell us the opposite. 

2. jalfi can not have been denoted by the kenning Rognir 
soknar, because nobody can detect in the words »the Othin 
of the battle» that I’jalfi, who is introduced here for the first 
time, is meant by them. It is likewise improbable that 
soknar at the beginning of the third line and Rognir at the 
end of the fourth belong together. 3. F. Jonsson makes the 
following construction: Rognir soknar vard fyrri i for med 
sagna svipti an arma farmr galdrs hapts. But if a poet wants 
to say: »The battle-Rognir went earlier with Thor than Loki» 
he can not express this by the words: »Earlier went Loki 
with Thor than the battle-Rognir.» I accept the inter¬ 
pretation of E. A. Kock x , who reads farmr arma meinsvarrans 
=Loki, obviously an imitation of the kenning farmr Sigvinjar 
arma in Haustlong. Then Loki is the subject of the clause; 
he is fyrri gorr i for than Rognir, or according to Kock than 
galdrs Rognir. Now Rognir is a name for Othin, and galdrs 
Rognir can be no other, as Othin is especially connected 

1 Notationes Norroenae § 2106. 


Jan de Vries, The problem of Loki 

FFC 110 

with galdr; the addition of this word is quite superfluous 
and it depends upon our interpretation of the remaining 
part of the stanza, whether we shall combine galdrs with 
the following word or with the preceding one. Kock reads 
as follows: med svipti sagna soknar hapts and explains it in 
this way: soknar hapts = Gunnar — gunnar. Gunnar sogn = 
viglid and then gunnar sagna sviptir = Thor. I propose the 
kenning: med svipti sagna galdrs soknar hapts with the 
following explication: soknar hapt = Gunnr, the name of a 
valkyrja. The galdr of the valkyrja is the battle and then 
again as Kock. The end of the comparative clause is conse¬ 
quently: fyrri . an Rognir. 1 The stanza tells us that 

Loki was more eager to go than Othin himself. 

In the following stanza Loki is again mentioned; here 
Thor is called bolkveitir Loka, which F. Jonsson translates 
as »the assistant of Loki», and I. Lindqvist more satisfactorily 
as »den omintetgoraren av Lokes svek». The first time 
t»jalfi makes his appearance is in st. 9 and here the poet 
does not use any kenning at all, but calls him by his proper 
name fcjalfi. I am of opinion that the poem presupposes a 
journey of the three gods Thor, Loki and J>jalfi. 

It is very remarkable that in the course of the narrative 
Loki is relegated to the background and I>j&lfi becomes 
more prominent. Still, if Kock is right in his interpretation, 
Loki is once more mentioned in st. 10, where we read: 

ogndjarfan laut arfi 
eiSsfjarbar hug meira; 
skalfa |>brs ne [naifa 
f>rottar steinn vid otta. 

Kock reads the word eids as eirs, which is required by the 
assonance ( eirs : meira) and explains it as follows 2 : fjardar 

1 H. de Boor seems to accept the same interpretation, cf. Deutsche 
Island-Forschungl930,1, p. 135 note 67. 

2 Notationes Norroenae § 455. 

FFC 110 Chapter V, Loki as the companion of Thor 


eir is »gold» and the heir to the gold is Loki. He considers 
this as an allusion to the tale of Andvari and argues that 
the symmetry of the stanza is much better, if the first half 
of the stanza tells about the three gods together ( slridkvid- 
jendr ), the second part about Loki, Thor and jalfi indivi¬ 
dually. There are,however, some objections. The parallellism 
with st. 21 

VreiQr sto5 Vrosku broQir, 
va gagn fa5ir Magna; 
skelfra f>ors ne t>jalfa 
ftrottar steinn vi5 otta. 

shows that it is not necessary, that the first part mentions 
Loki, because in st. 21 Loki is also altogether omitted. 
Moreover the allusion to the Andvara-myth is very question¬ 
able, not only because Loki can hardly be called the heir 
to the gold, but also on the ground that it is open to serious 
doubt, if the tale of Andvari has been current as early as the 
10th century. I believe that the arfi eirs fjardar means 
Thor, for he is the son of Jor5 and in the artificial language 
of this poet a kenning »the fjord of the ore» for »earth» is 
not at all improbable. 1 

I arrive at the conclusion that Eilifr tells about a journey 
to the giant’s home undertaken by three gods: Thor, Loki 
and frjalfi. In the Snorra Edda the last one is nowhere 
mentioned; here we find only Loki as the companion of Thor. 
The most striking contradiction is in the episode of the 
wading through the river Yimur. Eilifr says (st. 9): 

1 Cf. gammleid for »Loki», hlympd for »stick» a. s. o. The following 
kennings for earth may be compared: gaupu ver in st. 5 of the same 
poem; radyris vorr (Skj B I, p. 254, 14, 2). The earth is a sea, a fjord, 
but then qualified by something belonging to the earth. Radyris vorr 
is »the wake of the roe». Would a kenning »the fjord of the ore» 
be less appropriate? 

62 Jan de Vries, The problem of Loki FFC 110 

Unnz me5 fta sinni 
(aflraun var feat) skaunar 
a seilhimin sjola 
sjalflopta kom bjalfi. 1 2 

Snorri on the contrary: en Loki helt undir megingjardar. In 
this part of his narrative Snorri seems to make use of a 
poetical source, because 1. he quotes a fornyrdislag stanza, 
2. when he tells of the giant’s daughter Gjalp, who caused 
the river to flow so impetuously, he uses the alliterative 
formula at osi skal a stemma, 3. he makes an allusion to the 
help of the rowan-tree, again quoting a proverb reynir er 
bjorg f-ors? The question is this: has Snorri left bjalfi out 
on purpose, or has Eillfr added this god? For it is quite 
certain, that in the original myth Thor has had only one 
companion, because it is impossible that both Loki and 
i»jalfi could have taken hold of his waistbelt to cross the 
swollen river! Now Eillfr can not suppress the fact, that 
Loki has made the journey together with Thor; this is in 
accord with Snorri’s account; hence we may conclude that 
it was Loki, who originally accompanied Thor. 

Eilifr has added t>jalfi, not because, as Miss Gras sup¬ 
poses 3 , he was a fervent admirer of Thor and could not 
allow his hero to journey together with the low-minded 
Loki 4 , but because he knew from other myths that pjalfi 
was usually the god’s companion. He could not oust Loki 
from the introductory part of the tale, because he played a 
part here, exclusively his own; but later on during the 
journey he could replace him by J>jalfi: it does not matter 

1 Cf. for the interpretation E. A. Kock 1. c. § 1832. 

2 Cf. Feilberg, Bidrag til en Ordbog over Jyske Almuesmaal III, 
p. 123 s. v. ron. 

3 L. c. p. 39. 

4 The poet says in st. 3 that Loki is quite prepared to accompany 

FFC 110 Chapter V, Loki as the companion of Thor 


if this servant or Loki saved his life by clinging to Thor’s 
waistbelt. 1 

We may ask: what is the meaning of the myth? Thor and 
Loki journey to the giants and have here a difficult struggle. 
Moreover Thor seems to have gone without his usual weapons; 
the poem of Eilfir does not clearly state this, but Snorri is 
on this respect very positive. The introductory story serves 
to explain why Thor starts on this dangerous journey sva 
at hann hefdi hvarki hamar ne megingjardar. This is, however, 
not quite exact, for he tells himself that Loki helt undir 
megingjardar , while crossing the river Vimur. Snorri has a 
curious way of accounting for this surprising fact: he comes 
first to the house of G-rlbr and hon ledi honum megingjarda 
ok jdrngrei/pr er hon atti, ok staf sinn, er heitir Gridarvolr 
Indeed, Thor makes use of the last gift while wading the 
river and when he crushes the back of the giant’s daughters, 
who are concealed under his chair. But the fact that this 
gygr gives him the megingjardr does not agree with the 
introduction, Snorri himself has placed before the myth. 
Eilifr seems also to suppose that Thor has his magic belt, 
for st. 7 can hardly be explained otherwise: 

Harbvaxnar let harbar 
halllands of sik falla 2 
gatat njotr hin neytri 3 
niarb- rab fyr ser -gjarbar. 

1 If one wishes to argue as Miss Gras does, one may say that 

the poet had too high an opinion of Loki to represent him as a 
coward, who was dragged through the river suspended at Thor’s 
waistbelt; this was a role more fit for the servant Pjalfi. 

3 I adopt Kock’s text (NN § 449), although I do not agree with 
his translation. But I have not found a better solution. The text of 
Jdnsson seems to me impossible. 

1 I leave out the word madr, because it is superfluous in the 
metrical scheme. Is it perhaps a distorted dittography of niarb in 
the following line? 


Jan de Vries, The problem of Loki 

FFC 110 

|>verrir let, nema {>yrri 
f>orns barna ser, Mornar 
snerribloS, til svira 
sal{>aks megin vaxa. 

The exact interpretation of the stanza is doubtful 1 , but we 
may give the following translation: »Thor endured the 
fierceness of the river; the god of the belt did not find a 
useful way of defending himself, but then he said: if the 
rushing blood of the giantess does not end, I shall allow my 
strength to grow as high as heaven». I think it probable 
that he does this by means of his magic belt: then it is of 
importance that he is called in the preceding lines Njotr 
njardgjardar (according to Kock: Njotr gjardar). This ken¬ 
ning is used purposely and shows the god in possession of 
his belt which will now be necessary for him. 2 

Hence we may conclude, that Thor went to GeirroSr 
without his hammer. But when he has arrived in the giant’s 
hall and has played the play with the red-hot bolt, the poet 
proceeds (st. 19): 

Glaums niSjum for gumna 
gramr me8 dreyrgum hamri. 3 

Now, quite unexpectedly, he smashes the giants in his usual 
way with his hammer. It seems probable to me, that he 
gets his hammer in the hall of the giant, therefore quite 
the same idea as is expressed in the f>rymskvi8a. The 
parallelism between the two mythical tales goes even farther: 

1 Cf. E. A. Kock, NN § 449—450 and K. Reichardt, Studien zu den 
Skalden des 9. und 10. Jahrhunderts, p. 207 sqq. 

2 Perhaps of the kenning Marnar snerriblod the same may be 
said: the river is the blood of the giantess, because it is she who 
sends forth the gushing water. If this is right, then Eilifr has known 
the same myth about the giant’s daughter Gj41p, as Snorri relates. 

3 Cf. Kock NN § 466. 

FFC 110 Chapter V, Loki as the companion of Thor 


here as well as in the Eddie poem he goes with one companion 
and this companion is Loki. Snorri has not told the end of 
the tale, because with the death of G-eirro&r the most 
stirring part is over. He does not relate a myth for the 
sake of its religious meaning, but only as a queer interesting 
tale of former ages, by which obscure kennings may be 

I arrive at the following conclusion. The myth of 
G-eirrohr has been told in two different poems. First a poem 
in fornyrQislag with many proper-names and several details. 
Then in the skaldic poem of Eilffr, as usual hazy as to the 
exact form of the myth, but told with great skill in stylistic 
matters. In the original myth Thor went together with 
Loki to a giant, where he gets back his hammer after having 
smashed the trolls to atoms. Perhaps this myth itself is 
only a »mythological tale», built upon the well-known idea, 
that the thundergod sometimes has to visit the giants to 
regain his weapon. 1 On this material a poet could invent a 
tale, making use of motives common to stories about a fight 
with the giants. But this is of less importance for my purpose; 
I am content with the result, that in this mythical tale Loki 
is the companion of Thor. 

2. The myth of the giant-builder. 

In the Gylfaginning (ch. 41) Snorri tells a curious tale, 
how the gods made a bargain with a giant to build Asgardr 
for them. The castle had to be finished within one year; 

' 1 Haggarty Krappe has explained the wading of the swollen 
river as a myth about the arrival of the god of fertility in spring, 
carried by a sea-giant unto the land (Etudes de mythologie et de 
folklore germaniques p. 79—99). If this might be true, then the 
tale as it is actually told in Eilifr’s poem and in the Snorra Edda is a 
very distorted form of the original myth and has .hardly any more 
meaning than that of a motive. 



Jan de Vries, The problem of Loki 

FFC 110 

then the architect would receive as his reward the sun and 
the moon and moreover the goddess Freyja. With the help 
of his horse SvaSilfoeri 1 he gets on with such speed, that the 
gods are afraid of his success and they compel Loki to thwart 
him in his work. Loki changes himself into a mare and 
entices the giant’s horse to run away with him. The giant 
seeing that he is the dupe of the cunning of the gods becomes 
furious, but happily Thor arrives to smash him to pieces. 
After some time Loki foals a miraculous horse, the famous 

At the end of this narrative Snorri quotes two stanzas of 
the Voluspa, which tally with the tale as he records it and 
which he has even used in his paraphrasis of the myth. 
Scholars have drawn attention to several small incongruities 
in the tale, by which it seems to be proved that Snorri, as 
he was wont to do, has rearranged the myth into its present 
form. It will be sufficient to refer to the articles of the 
Swedish folklorist C. W- von Sydow about the Scandinavian 
folk-tale of Finn 2 , where a critical analysis of the Old-Norse 
myth has been given and the folkloristic material compared. 
He arrives at the following conclusion: the popular tale 
current about the building of different churches in Norway 
and Sweden has originated in Northern Scandinavia, most 
likely in Swedish Norrland (JamtlandP). Its original however 
is a heathen myth, of which we have a distorted form in the 
Snorra Edda. He is furthermore of opinion that Loki and 
his trick with Svadilfceri does not belong to the myth, but 
that Thor is originally the real hero of the story. By com¬ 
paring the tale of the Snorra Edda, the two stanzas of the 

1 See for the form SvaSilfoeri in stead of SvaSilfari: H. Pipping: 
Eddastudier II (Studier i Nordisk Filologi XVII, 3) p. 19. 

2 Studier i Finnsagnen och beslaktade byggmastarsagner, in the 
Swedish periodical Fataburen 1907, pp. 65—78, 199—218 and 1908, 
pp. 19—27. 

FFC 110 Chapter V, Loki as the companion of Thor 


Voluspa and the frame of the Alvissmal, he constructs the 
following primitive form: 

Thor (or the gods) have a building made by a giant. 
The work must be finished in three nights (or in one 
night) and the giant’s reward will be Freyja (or Thor’s 
daughter) and moreover the sun and the moon as well. 
The work proceeds very fast and then Thor delays the 
giant by a conversation, until the sun rises in the mor¬ 
ning and the giant (as well as his horse) are changed 
into stones. 

It can not be denied that this is a sagacious solution of 
the problem, but it is after all nothing but a mere hypothesis, 
which presents moreover several weak points. If we compare 
this original form with the Eddie tale, we find that the 
way, in which the giant is delayed, is quite foreign to the 
myth; it is borrowed from the Alvissmal, but von Sydow 
failed to prove that the frame of this poem has anything to 
do with this myth. It is a poor trick of the gods to stop the 
giant in his work by a simple conversation; what kind of con¬ 
versation they held, Von Sydow leaves it wisely undecided, 
Finally there is in this reconstruction no place for the horse, 
although it forms a prominent part in the heathen myth. 

If we, on the other hand, compare Von Sydow’s original 
form with the popular tales, we find no less important 
deviations. The giant is not delayed by a conversation, but 
he is frustrated in his work by the mentioning of his name. 
Von Sydow supposes the influence of an other folktale (the 
type of Titeliture), but then, if we compare other popular 
forms of the tale, that have not been influenced by the 
Titeliture-tale, we never find the solution of Von Sydow’s 
hypothetical tale. There is one more important difficulty. 
In the myth the giant requires as his reward sun, moon and 
Freyja x ; in the popular tale we find as his claim either the 

1 H. Pipping, Studier i Nordisk Filologi XVII (1926), 3, p. 81 
is of opinion that the gods promised to give celestial bodies and 


Jan de Vries, The problem of Loki 

FFG 110 

sun and the moon, or Saint Olaf, who wants the church to be 
built. Only if Olaf would be able to guess the troll’s name, 
the pact would be dissolved. 1 

Here is clearly an accumulation of motives. The guessing 
of the name, as belonging to the Titeliture-type, may be 
dismissed at once. But then there still remains a choice 
between the sun or moon on the one hand, on the other 
the person of Saint Olaf. This is illogical and moreover it 
finds no support in the tale itself. One might object that 
Olaf, although a saint, has not the power to dispose of the 
celestial bodies. In popular tales, found in Scandinavia as 
well as further abroad, the reward of the giant (or the devil) 
is the soul of the man who wants the building to be done. 
I am of opinion, that in the tale of the saint the motive of 
the sun and the moon does not belong at all, but that it is 
borrowed from an other tale, where the actor is a being, 
who is able to dispose of sun and moon. This is of course 
a myth, where a god gives the commission. Hence the 
special form of the Scandinavian tale about the giant- 
architect has been influenced by a myth; in this way we 
are able to account for the deviations from the typical form. 
This, however, does not imply, that- the tale as a whole 
goes back to the heathen myth; on the contrary we must 
try to find out, what may have been the form of the original 
tale, that lies at the bottom of the Eddie myth, as well as of 
the Scandinavian folk-tale. 

Now tales about supernatural beings, trying to make a 

that accordingly Freyja means the planet Venus. As this expli¬ 
cation is of no particular value for my discussion of the tale, I shall 
not enter here upon a criticism of this opinion, although I do 
not agree with it. 

1 Gf. besides the examples given by Von Sydow also Storaker- 
Fuglestvedt , Folkesagn samlede i Lister og Mandals Amt p. 100 
Nr. 138; Skar , Gamalt or Ssetesdal III, p. 41 and IV, p. 14; Finlands 
Svenska Folkdiktning VII, 1, p. 561. Also Bolte-Polivka I, p. 495. 

FFC 110 Chapter V, Loki as the companion of Thor 


palace, a road or a dike, abound in all parts of the world. 
They are commonly connected with natural phenomena, 
such as curiously shaped rocks or a row of stones through 
a lake; and as these only present an imperfect likeness to a 
well-finished edifice, usually the tale supposes that the spirit 
is thwarted in his efforts. As spirits do their work at night, 
the sunrise is the moment, when they must be ready; if one 
stone still remains to be put in place, the spirit has been 
unable to finish his task. In a tale from Wallonia the devil 
has to make a road in the course of one night; but when he 
came hurrying along with the last stone, he was surprised 
by the rising sun. He tumbled down with the rock and fled 
to hell, leaving the road unfinished and the stone with the 
imprints of his claws. 1 In stead of the rising of the sun 
itself a more poetic motive is sometimes chosen: the first 
cry of a bird, announcing the coming day. In a G-erman 
tale a devil makes a contract to build a mill: after midnight 
he fetched the miller to show him that the work was finished. 
But the man saw one stone wanting and immediately the 
devil flew away to fetch an other. But just at the moment 
he came back, the cock of the village Loffenau began to 
crow and in his fury the devil flung the stone at the mill, 
which was utterly destroyed. 2 

The tale becomes more dramatical, if the human being 
by his cunning succeeds in outwitting the devil. Both forms 
of the tale could be remodelled to this purpose. In a Javanese 
tale 3 a man must make a dike in a single night through 

1 Cf. G. Laport, Le folk-lore des paysages de Wallonie (FFComm 
Nr. 84) p. 66. This tale may be the truncated form of a fuller tale, 
where a man causes the night to finish earlier; but this is not likely, 
as the countless tales of giants turned into stones by their being 
surprised by the first rays of the sun prove that this a quite sufficient 
solution of the problem. 

2 Cf. Schreiber, Sagen aus den Rheingegenden II, p. 132 sqq. 

3 Cf. Knebel, Tijdschrift voor Indische taal-, land- en volkenkunde 
XXXVII, p. 499—500. 


Jan de Vries, The problem of Loki 

FF'C 110 

the river Senggarung. He begins the task with the aid of 
the spirits. At midnight the work is nearly finished; then a 
pandit begins to mutter prayers and succeeds in causing 
the sun to break through the clouds. The spirits are compelled 
to flight. 

The second form lends itself much easier to this remodel¬ 
ling of the tale. Then the farmer or his wife goes to the 
poultry-run and awakens the cock, who at once begins to 
crow. It is unnecessary to quote examples of this tale which 
is found in different parts of Europe. 1 It needs no further 
argument, that there are still more possibilities of preventing 
the spirit from finishing the imposed task. 2 

In these tales the human being, who accepts the aid of 
the devil, commonly has to pay for it with his soul. If the 
spirit succeeds the man must except to become its prey. 
This can of course be explained by the wellknown sacrifice 
of a human being on occasion of the foundations of a building 
or a bridge being laid. 3 As I had already the opportunity 
to point out, in the different variants of the Finn-tale the 
spirit requires the same reward; as a variant from Swedish- 
Finland expresses it very nicely, he must be rewarded »me 
m&naskin o mannablod». 4 

1 For Germany see: Miillenhoff, Sagen Nr. 412; for Belgium: De 
Cock-Teirlinck, Brabantsch Sagenboek Nrs. 232—238; for France: 
SibiUot, Le Folklore de France I, p. 378, II, p. 332, IV, p. 126; for 
Hungary: Ipolyi, Zeitschr. fur deutsche Mythologie und Sittenkunde 
II, p. 255. In another connection the same trick is found among 
the African Peoples of the Sudan, cf. Frobenius, Atlantis VIII, p. 119 
and 122. 

8 In a Mexican tale a man, who must build a house, gets the help 
of different animals. An animal, that shrieks very dreadfully, is 
compelled to cry. The llama’s, terrified by the noise, cast down their 
loads and so the task can not be finished (cf. Krickeberg, Marchen 
der Azteken und Inkaperuaner, p. 264). 

3 Cf. my paper in the Dutch periodical Volkskunde XXXII 
(1927), p 1—13. 

4 Finlands Svenska Folkdiktning VII, 1, p. 561. 

FFC 110 Chapter Y. Loki as the companion of Thor 


We have not yet succeeded in finding out the exact form 
of the tale underlying the Eddie myth as well as the Scandi¬ 
navian folk-tale; if we want to penetrate farther into the 
problem, we must now turn again to the form Snorri has 
given in his Grylfaginning. In his above mentioned paper 
Yon Sydow has made a thorough analysis of Snorri’s account, 
to which I may refer the reader; but I wish to discuss one 
question,which the Swedish scholar has not thought necessary 
to put:what is the exact relation between Snorri’s paraphasis 
and the two stanzas of the Voluspa? Nearly all who have 
given their opinion about this myth accept without any 
hesitation the fact, that in both traditions the same tale is 
meant, and accordingly they explain the enigmatical allusions 
of the Voluspa by the details found in the much fuller 
account of Snorri. 

At any rate the problem may be raised. In his account 
Snorri inserts a paraphrasis of the two stanzas of the Voluspa. 
By juxtaposition this will become quite evident. 

Snorra Edda Vgluspa st. 25 

I. (p. 42) Pa settusk go5in a domstola Pa gengo regin q 11 
sina ok leitubu raba ok spurbi hverr a rgkstola, 
annan, hverr pvt hefbi rabit at ginnheilog gob 

gipta Freyju 1 Jgtunheima eba spilla ok um pat gaettoz 
loptinu ok himinum sva at taka hverir hefbi lopt allt 

paban sol ok tungl ok gefa jgtnum. lsevi blandit 

eba aett i 9 tuns 
Obs mey gefna. 


En er aesirnir sa pat til viss 
at par var bergrisi kominn, pa 
varb eigi pyrmt eibunum ok 
kgllubu peir a P6r, ok jafn- 
skjott kom hann, ok pvi naest 
f6r a lopt hamarrinn Mjgllnir 

st. 26 

Porr einn par va 
prunginn mobi 
— hann sialdan sitr 
er hann sllkt um fregn! — 
a genguz eibar 
orb ok soeri, 
mal q 11 meginlig 
er a mebal foro. 


FFG 110 

Jan de Vries, The problem of Loki 

A few observations will not be out of place. 1 

1. The coming together of the gods to talk about the 
cause of the impending calamity is quite incomprehensible: 
half a page before we read: pa gengu cesirnir a tal ok redum 
radurn sinum ok var pat kaup gort vid smidinn , at hann skyldi 
eignask pat, er hann mcelti til . ..» The gods apparently knew 
very well how it came about that they were on the point of 
losing Freyja together with the sun and the moon. Ip their 
second discussion the gods come to the conclusion that Loki 
was again the cause of the mischief; it could hardly have 
been otherwise, as we have been duly informed that the 
gods had allowed the giant to make use of his horse, en 
pvi red Loki er pat var til lagt vid hann. So the gods need 
not come together to discuss a matter they ought to know 
all about. Snorri is too good an author to have made this 
bad arrangement by inadvertance; he has tried to fit in 
the stanza of the Voluspa on a place where it really does not 
belong. Moreover the stanza does not tell anything about 
the sun and the moon; Snorri has smoothed away the 
difficulty by paraphrasing the line hverir hefdi lopt allt Icevi 
blandit with the words: hverr /vi hefdi radii at . . . spilla 
loptinu ok himninum svd at taka padan sol ok tungl. But he 
makes two mistakes: 1 . Icevi blandit means charged with 
noisome venom 2 , not at all spoil the air by taking away 
the sun and the moon. 

2. The Voluspa makes allusion to something which has 
taken place at the moment the gods were holding their 

1 For further details I may refer the reader to the excellent 
criticism of Robert Hockert in his books »Vgluspa och Vanakulten» 
I and II. Perhaps it may be thought a reinforcement of my view 
upon this question that I came to my opinion about the relation 
between the stanzas of the Vgluspa and the tale in the Snorra Edda, 
before having read the argument of Hockert . 

2 I quote here the translation of Vigfusson and York Powell in 
their Corpus poeticum boreale I, p. 196. 

FFG 110 Chapter V, Loki as the companion of Thor 


council; in the Gylfaginning the disaster is only feared as 
being in the near future. 1 

I am convinced that Snorri has given a wrong interpreta¬ 
tion of the stanza. In his tale the giant will be rewarded in 
a double way 1. by the daughter of 6br and 2. by the sun 
and the moon. The latter condition belongs to the tale of 
the giant-builder, the former has been borrowed from the 
stanza of the Voluspa. Hence we are able to reduce the 
puzzling complexity of the giant’s reward into a more simple 
form, that moreover quite agrees with the popular tale 
about Finn. 

II. According to stanza 26 Thor as the only one of the 
gods fights the foes, obviously the giants (cett iotuns in st. 25). 
But by doing so all sworn oaths are broken. If this aims 
at the tale of the giant-builder, we are again confronted 
with a difficulty, which becomes painfully felt in Snorri’s 
analysis. The giant has already been thwarted in his plan, 
because Loki has deprived him of his horse, and consequently 
the giant has not finished his task. Hence there is no 
necessity whatever to break all these sacred oaths, the gods 
are freed from their promise. Snorri has felt the difficulty 
and he has tried to mend it, by saying that the giant became 
furious and menaced the gods. But then the gods were 
right in summoning Thor, because not they, but the giant 
himself had broken the pact between them. Snorri is a 
brilliant story-teller and he has contrived to make things 
as good as possible; but here the task was really too difficult. 
On the other hand the stanza of the Voluspa does not tally 
with the tale of the giant-builder. When I read the whole 
context, beginning with st. 21, I get the impression that 
after the serious trouble between the Vanir and the ACsir 

1 There is no possible way of escaping this contradiction; the 
remarks of S. Nordal, Voluspa, p. 63 only prove how desperate the 
situation is for those who wish to reconcile the Voluspa and the 


Jan de Vries, The problem of Loki 

FFG 110 

the question with the giant-builder is indeed too slight a 
matter to justify such exceedingly high-flown language as 
a gengoz eidar ord ok sceri, mdl oil megirileg er a medal fdro. 
This, however, is only a quite subjective feeling about the 
stylistic value of the expression; I do not wish to use it as 
an argument. This is only of importance for the interpretation 
of the Voluspa, to which I shall return presently; as to the 
tale of Snorri I think no more proofs will be required for my 
opinion that the stanzas of the Voluspa do not belong to it. 

The original form of the tale has become somewhat 
clearer. The giant proposes to make a castle for the gods; 
he asks as his reward the sun and the moon, naturally 
because he can keep them well enclosed in order that the 
giants may enjoy an everlasting darkness. But how has he 
been tricked? The folk-tale inserts the motive of the name- 
taboo, but this can not be correct as it adds a new element 
to the existing scheme and prevents the other ones from 
being developed in a satisfactory way. The Snorra Edda 
mentions the trick with the horse, which Von Sydow rejects 
as not original. But his solution that the giant is delayed 
by a conversation is so utterly devoid of epic character, 
that it must be rejected also. The question remains: does 
the horse belong to the myth or not? 

Snorri tells the tale on account of the horse Sleipnir. 
This is a fact, not sufficiently taken into consideration. The 
chapter begins with the question: »Hverr apann hest Sleipni 
eda hval er fra honom at segja?» And at the end it is told, 
that Loki is with young after his affair with Sva&ilfceri; then 
a colt is born, the famous Sleipnir. Snorri wishes to relate a 
myth about Sleipnir and from this standpoint we must 
judge the whole chapter. We may ask: what did he know 
about it? If we do not wish to suppose, that Snorri has 
drawn from inexhaustable mythical sources, but think it 
more probable that he has known on the whole the same 
mythological traditions, as have come down to us, the answer 

FFC 110 Chapter Y, Loki as the companion of Thor 


is not difficult to find. He must have known the HyndluljoS, 
as he quotes from it in his Edda, and here he could read in 
stanza 40: 

01 ulf Loki vi & Angrbo 5o 

en Sleipni gat vi5 Svadilfara. 

A tradition existed about Sleipnir being born from Loki, 
the other parent being SvaQilfari. Whether this was a horse 
or a human being is not clear from this stanza, but considering 
the character of Loki it was not impossible to presume that 
Sleipnir had been begotten by a mare, as one of his parents 
moreover must have been of horse-breed. But after all this 
single line of the short Voluspa was too meagre for a full¬ 
blown myth of Sleipnir. I suppose Snorri sought for a myth, 
in which Loki as well as a horse played a r6le. Evidently 
he found the tale of the giant-builder fit for this purpose 
which implies that here both were to be found. If we accept 
the opinion of Yon Sydow that Loki was subsequently 
added, we must at any rate imagine an original form of the 
giant-builder’s tale with the motive of a horse. Now in the 
popular forms of the Finn-tale a horse is not mentioned at 
all; if we conclude that in the original form of the Eddie 
myth the horse did not appear either, then we are confronted 
by the difficulty as to how Snorri hit upon the idea of 
introducing a horse into the story, where it had nothing 
to do and how he contrived to use this horse as one of the 
principal actors. It is for this reason that I am more willing 
to suppose the existence of a myth with a horse. 

Von Sydow has referred to an Irish tale, where a horse 
is found. 1 A Saint, St Mogue or Aidan, wants to build a 
church and he built it in the course of one night. He had 
the help of a grey horse, that brought the materials from a 
mountain. Now it is the devil who prevents the completion 

1 Fataburen 1908, p. 23. 


Jan de Vries, The problem of Loki 

FFC 110 

of the building. As von Sydow observes, the tale is out of 
joint, as devil and saint have changed their places. But we 
may infer from this variant, that in an original form the 
giant-builder made use of a horse. Did this horse play any 
r61e in the tale? Not necessarily, for the only reason for its 
being mentioned may have been that the rocks with which 
the folk-tale was connected suggested the form of a horse. 
This may have been the case in Ireland as well as in Norway; 
hence both tales may have originated quite independently. 
Then there has been in Norway somewhere a variant of the 
folk-tale, where the giant had the help of a horse, and it is 
only natural that in course of time this horse was also given 
an active part. 

As to the transition between myth and folk-tale, we 
grope in the dark. Surely Snorri has known a story about 
the building of Asgard; there is no reason to suppose that 
he had access to a poem, which now is irretrievably lost, for 
he does not quote any stanza from it, although he mentions 
two stanzas from the Yoluspa. He probably knew the myth 
from oral tradition. It must have told about the building 
of Asgard by a giant, who asked as his wages the sun and 
the moon. The giant had a horse to help him. How was he 
frustrated in his efforts? If we consider the different motives, 
used in this kind of tale, we will see that they do not fit in 
at all. The name-taboo is quite out of question. The trick 
of making the cock crow does not seem very appropriate 
for a story staged among the gods. Finally the giant being 
delayed till dawn is a motive without any epic value, if we 
do not hear in which way the delay was caused. Here, 
obviously, the horse may very usefully fill the gap in our 
reconstruction of the tale. 

Hence it was on account of the horse, that the giant 
failed to succeed. One of the gods, apparently has laid a 
trap for the giant by means of this horse. Was he Loki or 
Thor? Von Sydow supposes that it was the latter, because 

FFC 110 Chapter V, Loki as the companion of Thor 

he is mentioned in the 26th stanza of the Voluspa and he 
cheats a dwarf by means of a learned conversation about 
mythological problems. But if the connection with the 
Voluspa must be denied and if we do not accept the recon¬ 
struction of the plot of the Alvissmal as suggested by Von 
Sydow, we must take an other view of the question. Indeed, 
Thor plays no part whatever in the myth of the giant- 
builder; he first appears at the end, but here only as the 
result of Snorri’s combination with the Voluspa-stanza. 
So only Loki remains. And I must confess that the role 
of a cunning god, who by a trick entices the horse away 
from the giant, is much more in accordance with the charac¬ 
ter of Loki than of Thor. Loki, who is so well trained in all 
kinds of metamorphoses, changed himself into a mare in 
order to induce the giant’s horse to run riot in the woods. 
The birth of Sleipnir as a result of Loki’s trick is, however, 
an arbitrary combination of Snorri or of one of his pre¬ 

The tale, the contents of which I have tried to reconstruct, 
does not make the impression of a genuine heathen myth. 1 
First the question, how Asgard was built, arises more from 
curiosity than from real belief. Then the solution of the 
problem is rather burlesque; after all the gods were left 
with an unfinished castle. Hence it must have been a pure 
invention sprung from the brain of a mythographer. Then 
why not attribute it to the man who has recorded it? Indeed, 
this is not all impossible. Still I do not accept this solution, 
because it does not agree with the working method of Snorri, 
who is more of a compilator and an arranger of traditions 
than an inventor. If Snorri had himself composed the myth, 
only to explain the origin of Sleipnir, the name of a falsificator 

1 Cf. also K. Krohn, Ubersicht iiber einige Resultate der Marchen- 
forschung, FFCommunications Nr 96 p. 114—122, who arrives at 
nearly the same conclusion; whether the myth goes back to a Christian 
legend or to a popular tale, is of course difficult to decide. 


Jan de Vries, The problem of Loki 

FFC 110 

might be applied to him and it would not do credit to the 
earnestness of his intentions with regard to his handbook 
for poets. 

And finally, what is the result of our investigation for the 
appreciation of Loki? Very small, indeed. Loki is again 
the cunning god, appearing in the well-known role of the. 
man, who gives bad advice and afterwards has to remedy 
the dangers issuing from it. It is again the same Loki as 
in the myth of G-eirroQr or of I5unn; a figure more at home 
in a novellistic tale than in a real myth. 

Although the question as to how the two stanzas 25 and 
26 of the Voluspa must be explained, is of no importance 
whatever to the problem of Loki, the reader may expect, 
that an author, who rejects the commonly accepted explana¬ 
tion of their meaning, would present a new solution of them 
in stead of it. I will not exempt myself from this duty. If 
it may be considered as a reinforcement of my position, 
that I am not alone in my opinion, I may mention the 
suggestive book »Voluspa och Vanakulten» published by 
Robert Hockert. Still it can not be said that he has given 
a satisfactory solution, the more so as his general views 
about the meaning of the Voluspa are open to serious doubt. 
And in particular the idea, that in st. 25 the question hverr 
hefdi lopt allt Icevi blandit ought to be answered with »Othin», 
because he has induced the giants to work mischief by 
surrendering the goddess of fertility to them, is a suggestion, 
that is very ingenious, but still far from proved. 

The stanzas 25—26 are commonly considered as the 
consequence of the preceding ones. There has been a war 
between the Vanir and JEsir, in which the stronghold of the 
latter has been destroyed. Hence they want a new castle, 
which a giant promises to build for them, if they give sun, 
moon and Freyja to him. The sequence of the stanzas is not 
satisfactory and Nordal is inclined to suppose that one or 

FFC 110 Chapter V, Loki as the companion of Thor 


two stanzas are missing here 1 But there are other and more 
serious objections. How is it possible that the iEsir, after 
having been conquered by the Vanir, promise the VanaguQ 
Freyja as a reward for the giant’s undertaking? Are we to 
suppose that meanwhile the treaty between the Vanir and 
the iEsir has been concluded according to which Njor5r, 
Freyr and Freyja pass into the ranks of the iEsir? 2 If so 
who are the remaining Vanir? And how could Othin he in¬ 
clined to arouse the wrath of the Vanir afresh by giving up 
Freyja to the giants? 

I think these questions may suffice to show the host of 
difficulties meeting the generally accepted explanation. The 
real meaning of the preceding stanzas 21—24 is still obscure, 
in spite of the ingenious interpretations by Nordal and Van 
Hamel. 3 It is not even certain, that they really are connected 
with the war between the iEsir and the Vanir at all; the 
criticism of this view given by E. Mogk 4 is worth serious 
consideration. It is at any rate rather suspicious that the 
solemn lines about the gods united in council are repeated 

1 Voluspa p. 61. This view is accepted by Van Hamel, Arkiv for 
Nordisk Filologi XLII, p. 331. 

2 Cf. Ynglingasaga ch. 4. 

3 Van Hamel in the above-quoted paper combines st. 22 with 
the myth of Othin 's disappearance as told in ch. 3 of the Ynglingasaga 
(Cf. also the same author in Neophilologus XVI, p. 203). I consider 
this combination extremely doubtful. In the Ynglingasaga Frigg is 
simply taken by the gods Vilir and Ve (pa tdku peir ba&ir at eiga). 
This is not adultery at all, but, as Othin is thought dead, simply a 
case of levirate. This does not agree with Van Hamel's interpretation 
of st. 22 of the Voluspa, that Gullveig enticed Frigg to become the 
wife of her brothers-in-law during her husband’s absence. In the 
Ynglingasaga Othin accepts the state of affairs on his return; he 
simply takes his wife back without showing any resentment. In the 
Vpluspa Othin can not bear the insult and wages war, not against 
his brothers, but against the Vanir. — Cf. also H. Schtick, Studier 
i Nordisk Litteratur- och Religionshistoria II, p. 190 sqq. 

2 FFCommunicalions Nr. 51, p. 5. 


Jan de Vries, The problem of Loki 

FFG 110 

with such a very small space between (st. 23,1—4 and 
25,1—4). Moreover the stanzas 23—24 and 25—26 seem 
to form two parallel groups, each consisting of one stanza 
telling about the meeting of the gods, and a second one 
about a god who begins a fierce fight (Othin in st. 24, Thor 
in st. 26). Hence it is quite possible that they belong to 
parallel traditions and that only one of these groups originally 
had a place in the Voluspa. It can not be said beforehand 
if st. 23—24 or st. 25—26 are spurious. The line pal var 
enn folkvlg fyrst i heimi (st. 24, 3—4), a terrible »crux 
interpretations, may be adduced as an argument against 
the originality of st. 23—24. 

In the light of these considerations we may ask if it 
would not be better to arrive at an understanding of the 
stanzas 25—26 solely on the basis of their contents. We 
learn the following things: 1. The air has been poisoned by 
supernatural beings ( hverir in R is a reading, not to be 
dismissed as lightly as scholars are inclined to do), 2. Freyja 
is given to the giants. 3. Thor is the only one who dares to 
fight, 4. By doing so the most sacred oaths are broken. 

If we read these lines quite unbiassed, we are reminded 
of several wellknown Scandinavian myths. The giants are 
always seeking to possess Freyja. In ch. 17 of the Skald- 
skaparmal Hrungnir threatens to take Freyja and Sif away; 
the Eddie poem I>rymskvi5a is based upon the motive, that 
Freyja is to be married to a giant. 1 Obviously the meaning 
of this myth is the following: the goddess of fertility 
disappears during winter from the earth and goes to the 
realm of the giants. That means the temporal victory of 
these demons, who are now able to send snow- and hailstorms 
by which the air loses its propitious qualities for cultivation 

1 Although this poem is very recent (cf. my paper in the Tfjd- 
schrift voor Nederlandsche taal- en letterkunde XLVII, p. 251—322) 
the motive itself may be much older of course. 

FFC 110 Chapter V, Loki as the companion of Thor 


(cf. Icevi blandit 1 in line 25, 5—6). In spring-time Thor has 
the task of destroying the power of the giants; the mythical 
expression varies: he goes to fight the trolls, he goes to 
fetch back his hammer (f>rymskvi5a) or Freyja (st. 26 of 
the Voluspa). 

If such a myth is meant, the enigmatical stanzas of the 
Yoluspa become intelligible. There is only one detail, we 
have not yet explained: the terrible oaths broken by Thor’s 
attack. But this motive is again of wide occurrence: in fact 
he often seems to be willing to disregard the oaths of the 
other divinities ( hann sialdan siir er hann slikt urn fregn). 
It is related in the Lokasenna, that Loki only leaves the 
feast of the gods, because Thor threatens to slay him with 
his hammer. He calls his weapon here Hrungnis bani, and 
it is very remarkable that we find the same motive in the 
myth of this giant. In ch. 17 of Skaldskaparmal Snorri tells 
that Hrungnir, while persecuting Othin, comes in Asgard, 
where the gods invite him to drink. Of course grid must have 
been promised to him. 2 During the ensuing feast Hrungnir 
says that he intends to transfer Walhalla to the realm of the 
giants, to kill all the gods and to take Freyja and Sif to his 
home. The gods then call on Thor for help. He comes into 
the hall and asks why the giant sits at the table of the 
iEsir and who has given him grid. Othin has been compelled 
to do so, but Thor answers again at pess bods skal Hrungnir 
idrask , adr hann komi at. Hence obviously he does not 
intend at pyrma eidunum, and if in this case he does not 
swing the terrible Mjolnir, it is only because an other tale of 
Hrungnir has been added to this one. 

There are some striking similarities in the general 
structure of the tale between the Hrungnir-myth and the 

1 Cf. Hockert’s interpretation in Voluspa och Vanakulten p. 107. 

2 It is not said expressly but later on Thor asks hverr seldi Hrungni 



Jan de Vries, The problem of Loki 

FFC 110 

stanzas 25—26 of Voluspa. We may surmise, that a tradition 
has existed about Freyja’s sojourn with the giants and her 
deliverance by Thor contrary to all sworn oaths. Ordinarely 
the myths tell only about the danger of Freyja’s leaving the 
gods, apparently because in the mythological schematization 
Freyja always belongs to the principal deities. But the 
Voluspa is interesting because it represents Freyja really 
in the might of the demons. The idea of a regularly repeated 
sojourn with the demoniac powers, as is so clearly expressed 
by the Greek myth of Persephone, has not been preserved 
in Old-Norse tradition. It has been turned into a fact, 
that occurred only once and that accordingly must have 
taken place in a very remote past. It is difficult to say where 
it exactly fits into the mythical history of the gods, because 
the Old-Norse tradition never did arrive at a definitely 
developed system of cosmogony and eschatology. The poet 
of the Voluspa has made an effort in this direction. Perhaps 
the place of the stanzas 25—26, if it be accepted as the 
right one, may give us further indications about the proper 
meaning of this tale in connection with the other parts of 
the poem: it would be quite out of place to prolong this 
digression here, which has detained us too long in pursuing 
the subject of this paper: viz. the problem of Loki. 

3. Other myths of Loki as companion 
of Thor. 

In the humorous tale about Thor’s visit to see the giant- 
king Utgardaloki which Snorri tells at full length and with 
evident pleasure in his Gylfaginning 1 , Loki and fj&lfi are 
the companions of the thundergod. This story is a curious 
example of the so-called »Mythenmarchen», with special 
stress upon the second part of the word. C. W. von Sydow 

1 Ch. 43—46. 

FFC 110 Chapter V, Loki as the companion of Thor 


has made it the object of a thorough investigation, published 
in the Danske Studier 1 and he came to the very convincing 
conclusion that this cycle of tales was highly influenced by 
Celtic tradition. 2 This seems to me at any rate quite in 
accord with that part of the tale, where Loki is mentioned, 
that is, in the contests that take place in the giant’s hall. 
But for the eating-contest of Loki itself Von Sydow is unable 
to produce a Celtic example, and therefore he supposes, 
that this may be a quite independent product of a Scandi¬ 
navian poet’s fancy. 

Loki’s contest is the following: he is placed before a 
trough with meat, at the other end of which on the side of 
the giants Logi is placed. The one who succeeds in eating 
first half of the vessel, will be the winner. They meet each 
other in the exact middle of the trough, but while Loki has 
eaten only the meat, Logi has swallowed meat and bones 
and even the wooden vessel itself. Loki is ignominiously 
defeated, but he may have found consolation in the fact 
that his opponent was fire itself (villieldr as Snorri calls it). 

Logi as the opponent of Loki has given food to the 
supposition that Loki himself was a fire-demon as well. 
It is not necessary to discuss this opinion, as Miss Gras has 
lately made some good remarks about it. 3 But she goes 
rather far in her own conclusion, when she says: »if Loki in 
popular tradition had been accepted as a fire-demon, he 
could not have played this role, because in this case he 

1 Tors fard till Utgard, Danske Studier 1910, p. 65—105 and 

2 The opposite view has been defended by F. von der Leyen in 
Paul und Braune’s Beitrage XXXIII (1908) p. 382—390 and Deut¬ 
sches Sagenbuch I, p. 201, who says that the Eddie story was turned 
into a melancholy folk-tale in Ireland. A. Wesselski is of the same 
opinion, cf. his Versuch einer Theorie des Marchens p. 59. For rea¬ 
sons given below I accept the explanation of Von Sydow. 

2 Cf. 1. c. p. 46 sqq. 


Jan de Vries, The problem of Loki 

FFC 110 

could not have been the losing party». This holds good, as 
long as the source from which this is drawn, is a real myth, 
but when it is only a pleasant tale about gods and giants, 
the real proportions may be distorted. This author wants 
to show us the uncommon view of the gods being made 
utterly ridiculous by the giants. He even goes as far as to 
say that Thor can not lift the pseudo-worldsnake from the 
floor, although it is a well-known myth that he succeeded 
in fishing it up from the bottom of the ocean. When Thor 
in other poems, like the HymiskviSa and the f>rymskvi8a, 
pays a visit to the giants, his appetite is so extraordinary 
that the trolls are quite at a loss what to do. Here, however, 
he is presented with a goblet of mead and he only contri¬ 
ves to drink a nearly imperceptible quantity. In truth, 
only such an irresponsible joker as this author must have 
been, could have hit upon the idea of telling about a contest 
between the fire-god and a fire-demon, in which the former 
was defeated. 

The author chose this form of contest, because he knew 
of a connection between Loki and the element of fire. 
Obviously this makes the jest of the whole insipid story. 
It does not follow, however, that Loki has been a fire-god 
in popular tradition, nor that this was his original signifi¬ 
cance; we only can assume, that the combination of the 
words Loki and logi was so very close at hand, that such a 
disrespectful fellow as the author of this novellette has been 
could use it as an excellent pun. Now we understand why 
Yon Sydow could not find any parallel in Celtic tradition; 
this is a tale only to be imagined by a Scandinavian humorist. 

This author knew that both Loki and I^jalfi occurred as 
the companions of the thundergod. Of course he found this 
idea in other myths. But it seems that in his opinion l>jalfi 
was the real servent of Thor, for the contest of this god with 
the mind was only significant if pj&lfi was renowned for 
swiftness this being a good characteristic of a servant and 

FFG 110 Chapter V, Loki as the companion of Thor 


messenger. As to Loki lie obviously could not find anything 
in real tradition which suggested a contest with a giant; so 
he had recourse to the name, sufficient proof for the poverty 
of the traditions about Loki as the companion of the thunder- 

The same may be said about an other myth, where 
Thor and Loki together pay a visit to the giants. In the 
i>rymskvi5a a merry tale is told about their adventures. 
Scholars all agree in rejecting this farce as a real myth, 
although the poem itself is generally considered to be of 
rather high antiquity. The contradiction between these opi¬ 
nions does not exist for him who agrees with the present 
author in assigning this Eddie poem to a very recent date; 
for the numerous arguments in favour of this opinion I may 
refer the reader to my above-named paper on this subject. 
For the present it is only of importance to know where the 
poet got his material from. 

Loki flying in the falcon-dress of Freyja is a motive we 
are already acquiainted with (§ IV, 2). The formal way, 
in which it is lent , seems to denote that the poet tries to give 
a plausible explanation for this interchanging of divine 
attributes. At the foundation of this motive lies perhaps 
the idea of Loki being capable of flying about in the guise 
of a bird; this is not strange because he has the power to 
assume all kind of animal shapes. Making use of a machinery 
for flying sounds somewhat materialistic and gives to Loki 
the appearance of a godlike Egill. 

For the rest Loki is the cunning adviser of the gods, 
when they are in trouble. The contrast between the powerful 
Thor with his mighty force and his poor brains on the one 
side and the little, clever Loki on the other, is quite in 
accordance with the usual types of heroes in folk-tales. All 
mythical essence has evaporated from tales like this one, 
and they make more the impression of a »Marchen», trimmed 
up with some high-sounding hieratic names, than of an 


Jan de Vries, The problem of Loki 

FFC 110 

original myth, contaminated with or debased to a common 

This presents to us the interesting problem about the 
relation between the Old-Norse myths and the »Marchen». 
After the investigations of P. von der Leyen 1 it has become 
quite clear that in the tales of Snorri and Saxo G-rammaticus 
as well as in the Eddie poems, several reminiscences of popu¬ 
lar tales are to be found. But not in the exact form of the 
current tradition, but only as separated, often queerly 
altered and arbitrarely combined motives. The reason of 
this curious fact is formulated by v. d. Leyen in this way: 
the Old-Norse poets heard the tales told to them, but because 
they were a fresh acquisition to the literary stock people 
were not prone to admit them to the holy precincts of 
mythological traditions. And these tales were new, because 
they spread from Asia in the early Middle-Ages, especially 
from India, throughout Europe and of course arrived only 
relatively late on the shores of the Baltic and the North 
Sea. These tales made a profound impression upon the Nor¬ 
thern mind; they furnished new material for the adornment 
of worn-out myths, but the poets did not yet allow them 
to be treated on an equal footing with the old national 

This conclusion has been accepted by several scholars. 
Still it is merely a theoretical reasoning without any support 
in the real facts of Old-Norse literature. 2 What we know 
is this: we find in the mythological tales several motives, 
that occur also in the »Marchen», but the »Marchen» itself 
is never reproduced in its unaltered form. The explanation 
of this singular fact is a matter of interpretation. One might 
argue that the motives themselves belong to literary property 

1 Das Marchen in den Gottersagen der Edda (Berlin 1899). 

2 Cf. also the objections raised by E. 6■ Sveinsson, Verzeichniss 
islandischer Marchenvarianten FFC. No 83 p. XV. 

FFC 110 Chapter V, Loki as the companion of Thor 


of most peoples and that it is not necessary to assume any 
influence resulting from the ocean of stories that continually 
poured over the frontiers of India. Hence they might be 
the result of what Bastian called the »Volkergedanken». 
And I do not think it unlikely that we might be compelled 
by future investigation to be more liberal towards the 
national literature of the European peoples. Moreover, if a 
motive properly belongs to a »Marchen», it does not follow, 
that it reached Scandinavia first about the eleventh century 
or even later. The Indian theory in its extravagant form 
has long been seriously shaken in its fundaments; »Marchen» 
may be composed even by other Indo-European peoples, 
they may be the property of the Germanic tribes as well. 
Each case has to be considered separately. 

But even if we a d mit, the thesis of v. d. Leyen that India 
is the Klondyke of folk-tales, we are not compelled to follow 
his argumentation with regard to the mythological tales of 
the Scandinavians. People objected to admitting them 
unaltered into the old national traditions, because they 
were so very recent. I do not see the value of this argument. 
If they were unknown and interesting, why did the poets 
not retell them, adapted to the sphere of divine beings, 
where things of magic and wonder were not out of place? 
We might argue in the opposite way of v. d. Leyen: the 
poets did not accept them unchanged because they were 
common property; the public would too easily detect the 
real source of the poet’s fiction. But if he only used some 
motives, placed in new surroundings or with some slight 
modifications, the reader might have got the agreeable 
impression of things unknown, that still at the same time 
were not quite foreign. If we argue in this way we could 
consider the curious appearance of folkloristic motives in 
Scandinavian mythology as a proof for the opinion, that the 
»Marchen» were widely current in the early Middle-Ages. 

But this conclusion would be as unjustified as that of 


Jan de Vries, The problem of Loki 

FFC 110 

v. d. Leyen. The mythological tales could not embody folk¬ 
tales in their unaltered form, because they are incompatible 
literary forms on the ground of their totally different 
character. We observe the same phenomenon elsewhere as 
well: as soon as the »Marchen» is introduced into higher 
literature, it must be remodelled and disintegrated to 
furnish material for unbiassed poetic fiction. This is but 
natural. The folk-tale is a complete scheme of action 
composed of folkloristic material, that is moulded into a 
solid form. It may be liable to contamination with other 
folk-tales; motives may be interchangeable; still on the whole 
the type of a folk-tale is relatively constant. Modifications 
are possible only in the sphere of folk-tradition. If the hero 
of the tale does not venture himself outside the world of 
wonder and chance, that constitutes the real scene of the 
»Marchen», it does not matter what kind of adventures he 
undergoes nor in which order they occur. 

As soon as the folk-tale is admitted in higher forms of 
literature its shortcomings are evident. The necessity of 
psychological treatment of the persons, the craving for 
logical connection of the facts, that are the real distinctive 
marks of all higher literature are incompatible with the real 
character of the folktale; it can not but wither in this chill 
wind of stubborn reality. »This succession of facts can not 
remain unchanged, such a stupid conception of the chief 
actor is incomprehensible to a more cultivated mind», in 
such a way an author might argue, who wished to annex 
the variegated charms of the folk-tale to his proper domain. 
The only issue was to annihilate the folk-tale itself, in order 
to make free use of the constituent motives, which might 
be adaptated to any possible story. 

The'fact, that the later mythographers of Iceland were 
unable to insert complete folk-tales into the traditions about 
gods and demons, is only a proof of the vigour and briskness 
of this traditional literature; only when the old spirit is 

FFC 110 Chapter Y, Loki as the companion of Thor 


quite broken down and people consider myths to be some 
specific kind of wonder-tales, time has come to confuse 
completely two literary forms, that belong to different 
plans. The attitude of the Scandinavian mythology towards 
the »Marchen» can not furnish us with any indications about 
the period when the wandering folk-tales were introduced 
into Scandinavia. 1 

After having written these remarks about the relation 
between the Eddie myths and popular tales I had the 
opportunity to read the highly stimulating book of Wesselski 
»Versuch einer Theorie des Marchens». He defends the thesis 
that it is the myths that are original and that consequently 
the folk-tales are derived from them. Generally speaking 
this view may be accepted, but even then it remains quite 
possible that mythological stories, occurring in later literary 
tradition, are deeply influenced by popular tales, which in 
their turn may again be composed of motives detached from 
original myths. This seems to be particularly the case with 
the tale of Thor’s journey to Utgardaloki, which in its 
character is not at all popular: although it makes use of 
different current motives from mythology as well as from 
story-telling, it has such an allegorical character, that it 
must be considered in its present form as a purely lite¬ 
rary work. The question remains then unsolved, whether 
this author, got his material from Irish traditions or the Irish 
tales are themselves derived from the Icelandic story. The 
latter view seems to me rather improbable, because the 
Eddie tale belongs to a relatively late period of myth-making 

1 It is also worth consideration that with actual myth-borrowing 
the disintegration of the tale is a very common feature; cf. P. Radin, 
Literary Aspects of North American mythology in Canada Depart¬ 
ment of mines, Geological survey, museum Bulletin No. 16 (Ottawa 
1915) p. 49—50: we have always to bear in mind that borrowing is a 
selective process. The problems connected with myth borrowing 
thus assume a far greater complexity than we are apt to give them. 


Jan de Vries, The problem of Loki 

FFC 110 

viz. to the eleventh or even twelfth century, when the rela¬ 
tions between Ireland and Scandinavia had become much 
less intensive, hence the opportunity for such a transfer of 
an Icelandic mythological story to Ireland seems to be 

4. Loki as Thor’s opponent. 

In chap. 33 (35) of the Skaldskaparmal Snorri tells a 
curious tale about the hair of Sif. Loki has cut it off, out 
of pure malevolence. Thor is very angry about it and com- 
pells him to go to the svartdlfar to get golden hair instead. 
The dwarfs Ivalda-synir are so complaisant to make besides 
the hair also SkiQblaQnir and Gungnir. Then Loki made a 
bet with two other dwarfs, that they would not be able to 
forge as splendid objects. They, however, begin successfully 
and make Gullinborsti and Draupnir. Now Loki, afraid that 
they will win, transforms himself into a fly and stings the 
dwarf, who handles the bellows. Indeed, the hammer 
Mjollnir is not quite faultless, for the shaft is too short. Still 
the gods, asked to give their verdict, consider the second set 
of objects superior to the first set. The dwarf wants to take 
the head of Loki, but he escapes. Thor, however, brings 
him back and then Loki says, that if the dwarf has the right 
to take his head, he must not touch his neck. Finally the 
dwarf sews the lips of Loki together. 

This tale belongs to the most difficult problems of 
Scandinavian mythology. We have no parallel traditions 
whatever and we are thrown on an analysis of Snorri’s text, 
which offers different puzzling contradictions. It is obvious, 
that this tale is a stew of different motives, as already 
F. Ohrt has pointed out. 1 There is no logical coherence 
between the different parts. The dwarfs are asked to make 

1 Hammerens lyde — Jsernets last in Festskrift til Finnur Jonsson 
p. 294—298. 

FFC 110 Chapter V, Loki as the companion of Thor 


golden hair for Sif; why do they forge three objects? What 
is the reason of the competition between the two groups of 
dwarfs? Loki succeeds in thwarting the dwarf, how is it 
possible that the object, made under such circumstances, 
er beztr af ollum gripunum? What is the reason of Loki’s 
escaping, followed by his being brought back by Thor? 
The shrewd remark of Loki, that the dwarf may not take 
hold of his neck, is a blunt motive, because Loki is not 
freed by it, but must subject himself to the sewing together 
of his lips. What is the meaning of this motive, to which no 
allusion can be found in any other literary work? 

If Snorri’s tale is composed of different originally inde¬ 
pendent motives, we must try to find the components. My 
attempt to make an analysis of this myth must needs be 
hypothetical; still it may be useful for a better understanding 
of the difficulties that face us here. The motive for Snorri’s 
telling about this myth is the question, why gold is called 
haddr Sif jar. In fact this kenning is nowhere to be found in 
Old-Norse poetry; the only case that may be compared with 
it is a stanza of Bjarkamal, where we read Sif jar svardfestum 1 
but this stanza belongs, as we already had the occasion to 
observe (cf § IY, 3) to a very late interpolation. 2 Has Snorri 
found the kenning haddr Sif jar in a poem, that is lost to 
us, or has he himself substituted the simpler word haddr 
for the poetical term svardfestar? And what is the meaning 
of this expression? The idea first enounced by Karl Wein- 
hold 3 that the hair of Sif means »vegetation», is not at all 
improbable. Perhaps it is better to modify it in this way, 
that the golden hair is a mythical connotation of the ripe 

1 F. Jonsson , Skjaldedigtning I B, p. 171, st. 5. 

2 See also A. Olrik , Danmarks Heltedigtning I, p. 99, who dates 
the stanzas in the 12th century. 

3 Die Sagen von Loki, Zeitschrift fur deutsches altertum VII 
(1849) p. 38. 


Jan de Vries, The problem of Loki 

FFC 110 

corn-ears, as J. G-. von Hahn has suggested. 1 At any rate 
the much simpler explanation, proposed by Miss Gras, that 
it only means the beautiful golden hair of the Germanic 
women 5 is quite insufficient, as it does not explain, why 
the kenning mentions the hair of Sif and not of any other 
goddess. 2 * * If Sif’s Fair means »vegetation», cut off in harvest 
time and reappearing in spring, Loki is not a fire-demon, 
as the older explanations suggested, but more likely a 
chthonic deity, the relation between him and Sif being 
comparable with that between Hades and Persephone. But 
such a character of Loki is elsewhere unknown and we must 
leave the possibility open, that Loki originally has nothing 
to do with the myth of Sif’s hair, but has been introduced 
afterwards to achieve the combination with the contest of 
the dwarfs. 

Two groups of superhuman smiths each forge three 
mythical objects. These quite correspond to each other: a 
weapon of a god (Gungnir ~ Mjollnir), a property of Freyr 
(Ski5bla5nir ~ Gullinborsti), an object of gold (Draupnir ~ 
Sif’s hair). We must conclude that this parallelism is inten¬ 
tional as there are no reasons for it in the character of the 
mythological objects themselves. Hence it is probable that 
one of the sets is original and the other is composed as its 
counterpart. Now the group Draupnir — Gullinborsti — 
Mjollnir is a logically correct triad, as it consists of important 
objects belonging to the three principal gods. This is not the 
Case with the other group, where the hair of Sif strangely 
appears besides Skidbla&nir and Gungnir. Moreover in this 
case we may find the reasons for the combinations. If a 
myth-teller wished to compose this new triad as an analogy 
to the other one, and if he wanted to combine it with the 

1 Sagwissenschaftliche Studien (1876) p. 131 sqq. 

2 R. M. Meyer, Altgermanische Religionsgeschichte p. 306 says 

rightly: »Mythisch bedeutsam miissen ihre Haare wohl jedenfalls 


FFC 110 Chapter V, Loki as the companion of Thor 


story about Sif’s hair, his argument might have been the 
following. The hair of Sif, as an golden object, answers to 
the ring Draupnir of the pre-existing group. Gnllinborsti 
leads of itself to the other famous property of Freyr Ski5- 
bla&nir, about which st. 43 of G-rimnismal says. 

ivalda svnir gengo i ardaga 

SkiSbladni at skapa 
skipa betst skfrom Frey 

nytum NiarSar bur. 

Here our mythographer found the name of the dwarfs, who 
made this precious object. Finally the third thing had to 
correspond with Thor’s weapon; the spear of Othin was not 
difficult to hit upon. 

In my opinion this is a possible solution of the problem. 
If we try the other way, we have to face much greater 
difficulties. For if the group Draupnir — G-ullinborsti — 
Mjollnir is secundary, it is inexplicable that this forms a 
logical combination of precious divine things. Moreover, 
how did he find the names of the dwarfs Sindri and Brokkr? 
The first one occurs once in a stanza of the Voluspa, where 
it is said that he has a golden hall a Nidavollom (st. 37), 
strangely misinterpreted by Snorri as if Sindri were the 
name of the hall itself (Gylfaginning ch. 51). Brokkr is 
quite unknown. Hence there may have been a tradition 
about the origin of the principal accessories of the three 
chief gods, in which the dwarfs Brokkr and Sindri were 

But this is not sufficient to explain the details of the 
contest. These are mostly well-known folkloristic mo¬ 
tives, although curiously distorted and even devoid of their 
original meaning. Loki as a fly appears in the myth of the 
Brisingaman, to be treated subsequently. His stinging of 
Brokkr has no effect, as the weapon of Thor is nonetheless 


Jan de Vries, The problem of Loki 


excellent in the war with the giants, and because Sindri and 
Brokkr win the game. This is the reason why I can not 
see the similarity with the Finnish magic formula about 
the origin of »iron», pointed out by Ohrt. Here we read in a 
variant from Pohjanmaa 1 : 

Herhilainen Hiien lintu 
katsovi katon rajasta, 
ruotehesta ruijottavi. 

Kantoi kaarmehen kahyja 
rauan karkaisuvesihin. 

This is not quite clear to me. What is the relation between 
the wasp and the snake-poison? Is this rhyme not rather an 
arbitrary combination of two different conceptions: 1. the 
idea of snake-poison in a weapon 2 and 2. the idea of the wasp 
as the devil’s instrument? 3 Besides the correspondence 
with the Old-Norse myth is meagre. Here the fly stings the 
smith, but the wasp drops the poison into the tempering- 
water. The effect is also quite different; on the one side the 
slight fault in the hammer’s shaft, on the other side the 
dangerous effect of iron weapons anf tools. I do not think 
that it is necessary to suppose any relation between the 
Finnish and Scandinavian traditions. In the Old-Norse 

1 K. Krohn , Suomalaiset syntyloitsut p. 81 (translation in FFC 
52 p. 85). 

2 Cf. Brot af Sigurdarkvida st. 19: eldi voro eggiar utan gofvar, 
en eitrdropom innan fadar. Halfdanarsaga Eysteinssonar FAS III, 
p. 543. Also Beowulf 1459 ecg wees Iren , atertanum fah (and cf. 
Klaeber’s note to this line, to which may be added my StudiSn over 
Faerosche balladen p. 12 sq.). See also Reichborn-Kjennerud , Var 
gamle Trolldomsmedisin (Vid. Akad. Oslo, Hist.-Fil. Kl. 1927 Nr. 6) 
p. 90. 

3 K. Krohn , Suomalaiset syntyloitsut p. 83 (FFC 52 p. 88) and 
0. Dahnhardt , Natursagen I, p. 165, 167, 188 sq, 334. 

FFC 110 Chapter Y, Loki as the companion of Thor 


myth the motive of Loki as a fly is a simple borrowing from 
the tale of the Brisingamen. 

The shrewd answer of Loki may he a wandering motive, 
like the wellkown story of Shylock. For the sewing of Loki’s 
lips I do not know any parallel; it may he suggested by the 
sharp and dangerous tongue of the caustic god. 1 

After this short discussion of Snorri’s tale the impression 
of a wildly fantastic combination of motives can only be 
strengthened. Obviously Loki is the central-point in it. 
His character is not very clear-cut. He is a trickster, amusing 
himself in cutting the hair of a goddess. But he is compelled 
to repair the mischief, that he has done; we met with the 
same conception in the myths of I9unn and the giant- 
builder. He has some connection with the dwarfs; hut only 
very superficially: he is a messenger of Thor to these super¬ 
natural smiths and he succeeds in making a quarrel between 
two groups of them. It seems rather hasardous to conclude 
on the basis of this myth, that he belongs in some way or 
other to the dwarfs. 2 Finally he outwits the dwarfs by a 
sudden disappearance (frustrated again by Thor’s inter¬ 
vention) and by a clever subterfuge. After all he is punished 
for his malevolence in a rather ridiculous way. This is all 

1 Cf. Miss Gras p. 69. We are reminded of the story told by 
Florus (Epitome II, 30) about the defeat of Varus: tandem, vipera, 
sibilare desisti. 

2 As Miss Gras p. 66—67 is inclined to accept. F. Graebner in a 
paper on »Thor und Maui» in Anthropos XIV—XV, p. 1106 compares 
Loki’s defeat as patron of a family of dwarfish smiths with the 
Indian myth of Tvastar and the Rbhus. This article, however, is a 
warning example against an ethnological method, that indulges in 
wild combinations and sweeping assertions without taking the trouble 
to ask what the real value of a literary tradition may be. If such a 
method pretends to give the clue to mythological problems, we can 
only emphatically reject it as superficial and misleading. But we 
should not judge the »kulturhistorische Methodeo by this specimen 
of its application. 


Jan de Vries, The problem of Loki 

FFC 110 

of no importance whatever and I see no reason to ascribe a 
particular value to these newfangled myths and even to 
give them the preference above the older sources, as Miss 
Gras wishes to do. 1 

There is one remarkable circumstance: Loki is unfriendly 
towards Thor. First he offends the wife of the thundergod 
by cutting off her hair, and secondly he succeeds in damaging 
the weapon of Thor. It is seemingly a mere chance that the 
fault was of such slight importance; if Loki had had his 
will, the hammer might have been quite useless. I do not 
think that this conception is original; it is at any rate 
more probable that it arose from the contrast between 
the formidable god of muscular strength and the clever 
representative of mental acuteness. This might develop in 
course of time into a relation between them, not unlike that 
of the human hero and the clumsy giant as is found in seve¬ 
ral folktales. 

The same character of Loki is found in two stanzas of the 
Hymiskvida, usually considered as a recent interpolation 2 : 

37. Forot lengi 
hafr Hlorri&a 
var skser skokuls 


en Im inn laevisi 

a&r liggia nam 
halfdaubr fyrir; 
skakkr a beini 
Loki um qlli. 

38. En er heyrt hafiS — 
hver af hraunbua 
er hann baebi gait 

hverr kann um £at 
gorr at skilia? — 
hann laun um fekk 
born sin fyrir. 

The last stanza supposes the same myth as Snorri tells in 
ch. 44 of the Gylfaginning 3 ; only here the children of a 

1 Ibid. p. 66. 

2 Gf. R. C. Boer , Die Edda II, p. 94. 

3 Cf. Gering-Sijmons , Edda-Kommentar I, p. 274. 

FFC 110 Chapter V, Loki as the companion of Thor 


peasant, not of a giant, are mentioned. We do not know 
which is the right reading. In Snorri’s rendering of the myth 
Loki is not mentioned at all; is it a self-evident conclusion 
from the fact that he accompanies Thor on the journey, 
when they have the adventure in the peasant’s home? 
Or has later tradition added this motive, because Loki was 
then considered to be the mischief-maker par excellence? 
It is curious, that the term inn Icevlsi Loki, also mentioned 
in a superfluous line of st. 54 of the Lokasenna, is used in the 
introduction of the tale about Sif’s hair: Loki Laufeyjarson 
hafdi fiat gort til Icevisi, at klippa har alt af Sif. Hence we 
may surmise that these mythological tales belong to the 
period when Loki had developed into a foe of gods and 

The stanza of the Hymiskviba tells about people who 
might know more about it; they are called by the technical 
term godmolugr. This is a very interesting detail. We learn 
from it, that there have been persons, who made a profes¬ 
sion of relating and probably also of composing myths. 
Hence we might ask, if they have left no traces of their 
activity in literary tradition. Now when we try to give 
account of the origin of ch. 33 of the Skaldskaparmal, we 
get the impression that it is a combination of different 
myths, possibly even partly only a later invent’on. But 
obviously Snorri is not its author. For he might be an 
arranger of the mythological material even in many cases 
freely handling, remodelling and combining the tradition, 
but he could not have invented this incoherent mass of 
motives, only to explain the kenning haddr Sifjar. Of 
course he knew the tale and inserted it here at full length, 
because he found this the proper place for it. 

But then we get a glimpse of the activity of a myth- 
maker, who arranges traditional material and invents new 
stories, clearly to divert a public, desiring to hear curious 
tales about the gods. This gobmolugr man may belong to 



Jan de Vries, The problem of Loki 

FFC 110 

a whole group of myth-mongers, who have had the important 
merit of preserving and handing down the heathen traditions, 
but who have not allowed them to pass unaltered through 
their hands. This class of mythographers includes serious 
traditionalists as well as mere inventors of mythological 
fables. They have apparently been one of the foremost 
causes of the self-contradictory, highly divergant conceptions 
to he^ found in Old-Norse traditions. And this is once more 
an opportunity to warn against the over-estimation of the 
younger traditions and the tendency to correct or interprete 
the older sources by the help of the later accretions. 

In this way we are able to explain, why we find in Snorri’s 
material such an enormous amount of mythological names, 
quite unknown in other sources. I already mentioned the 
names of the dwarfs Brokkr and Sindri. Still more curious 
is the fact that the thread, with which Loki’s mouth is sewn, 
has its proper name Vartari. We are reminded of the 
analogous fullness of detail in the myth of Fenrir’s binding; 
in ch. 33 of the Gylfaginning Snorri gives the names of 
LoeSing, Dr6mi and G-leipnir for the successive fetters of 
the monster. Obviously it is not Snorri who invented these 
names. But it is a common feature of the later development 
by mythographers, that particular names are given to the 
smallest details in the mythological tales. In this way their 
spurious inventions acquire the deceptive appearance of 
unimpeachable trustworthiness. Hence, if we meet with 
such suspicious details in the work of Snorri, we must be 
careful in making use of them for our conclusions. It would 
be unwarranted to reject them as mere inventions of his; 
on the contrary he found them doubtless in tradition. But 
this tradition itself was a queer medley of disparate material. 
For Snorri did not draw only from the pure sources of the 
poetic Bdda and the Skaldic poems; he made a still more 
extensive use of oral traditions, that had passed during two 
or three centuries through the hands of godmolugir men. 

FFQ 110 Chapter V, Loki as the companion of Thor 


We will find more instances for their activity later on in the 
tales about Loki and it seems to me that this fascinating 
god has had a particular attraction for the mythographers 
of the period, when men believed no longer in myths, but 
thought them fit for harmless amusement. 

5. Loki as the servant of the thundergod 
a. Exposition of Oink'$ view 

Hitherto we have dealt with those myths, that show 
us Thor and Loki accompanying each other on different 
expeditions against the giants. In some instances j>jalfi 
takes the place of Loki, and clearly as a result of secundary 
combinations both f>jalfi and Loki appear as the companions 
of the thundergod. As Loki only occasionally figures in 
this role and on the contrary J>jalfi is nothing else but the 
servant of Thor, we may infer that the latter originally 
belongs to the sphere of Thor’s activities, while on the other 
hand Loki has for some reason or another been associated 
with the thundergod in a function like that of bjalfi. Still 
the question may be raised, if it is not possible that one of 
the constitutive elements of the highly complicated figure 
of Loki is his having been exactly the companion and servant 
of the thundergod. 

It is Axel Olrik who put the question most energetically 
and proposed a brilliant solution. In a paper on this sub¬ 
ject 1 he collected a great deal of folkloristic material drawn 
from Scandinavian, Lappish and Bsthonian sources, all 
pertaining to the mythic conception of the thundergod’s 
servant. The representations on Lappish magic drums as 
well as a Swedish folktale led him to the conclusion, that 
a very original form of this conception was to be found 
here. The companion of Thor has neither a prominent 

1 oTordenguden og hans dreng» in Danske Studier 1905 p. 129-146. 


Jan de Vries, The problem of Loki 

FFC 110 

position nor a clear-cut character; he is completely dwarfed 
by the god himself. With this companion, whom Olrik 
considers to be especially Swedish, he connects the name 
of t>j&lfi, because it is more common in East-Scandinavian 
than in either Norwegian or Icelandic tradition. 

On the other hand there are some Esthonian and Finnish 
folk-tales about the thundergod and his servant, where 
the latter shows striking similarities to the figure of Loki; 
here we find, according to Olrik, reminiscences both of the 
Hymiskvi&a and the I>rymskvida. Hence Loki as the com¬ 
panion of Thor was known to the Teutonic peoples who 
have influenced the tradition of the Esthonians; he for¬ 
mulates it in this way, that these myths are an early borrow¬ 
ing from the most eastern »gothic» tribes. 

In Danish scholarly literature »gothic» is used as a general 
name for all Teutonic peoples 1 ; hence we do not know 
exactly what Olrik meant by this vague term. If the borrow¬ 
ing has taken place as early as the beginning of our era, 
the only tribes that can be taken into consideration are 
those belonging to the East-Teutonic stock (the Aestii in 
the days of Tacitus 2 , probably closely related to the Gothic 
tribes); they came, according to their own traditions from 
Southern-Sweden, but developed afterwards into a quite 
independant group of Teutonic peoples. In later times 
the Teutonic population of Esthonia was undoubtedly 
Swedish. Hence it seems to me that we have to take those 
»most eastern gothic tribes» either as identical with or as 
closely related to the Swedes. 

The distinction between a specially Swedish l>j£lfi and 
an »eastern-gothic» Loki gives the impression of being 
purely theoretical and even arbitrary. If we look at the 
Scandinavian traditions themselves, we come to the con- 

1 Cf. G. Schiitte, Vor Folkegruppe Gottjod I, p. 45. 

■ * T. E. Karsten, Germanisch-finnische Lehnwortstudien p. 206 


FFC 110 Chapter V, Loki as the companion of Thor 

elusion, that Loki has been known particularly among the 
West-Scandinavian peoples where he is an important figure 
in mythological traditions. The most outstanding conclu¬ 
sion is that Loki is especially Western, whereas pjalfi may 
have been originally more particularly Eastern. To t his 
logical observation, made by H. Celander 1 , Olrik violently 
objects 2 by stressing emphatically that he called J»jalfi 
Swedish and that the myths of Thor-Loki have their most 
original form in the countries beyond the Baltic. 

This, however, is only escaping from the necessary con¬ 
sequence of his own statements. It is indeed highly im¬ 
probable that there have been at the same time and with 
the same people two different mythical forms of the servant 
of the thundergod. If jalfi is Swedish, then Loki, who is 
prominant in West-Scandinavian tradition must be Nor¬ 
wegian-Icelandic. In the course of later development they 
penetrated into each others domain; hence j>jalfi has been 
adopted in Eddie mythology and on the contrary Loki was 
introduced into Swedish popular belief. But then the know¬ 
ledge about Loki among the most Eastern Gothic peoples 
becomes very enigmatical and we are obliged to try to make 
a clear distinction between the two different forms of one 
single mythical conception. 

Olrik’s view on the relation between the Esthonian 
myth and Scandinavian mythology of the thundergod’s 
servant has not been unchallenged. In the above men¬ 
tioned study Celander has contested the correctness of 
Olrik’s argumentation. I myself have expressed my doubts 
about it . 3 A new treatment of this problem is inevitable, 
particularly since we now have occasion to obtain more 
exhaustive information about the Esthonian tradition, 
than was possible in the days when Olrik wrote his study. 

1 Lokes mytiska ursprung p. 90. 

2 Danske Studier 1912 p. 91. 

3 Tijdschrift voor Nederlandsche Taal- en Letterkunde XLVII, 
p. 294. 


Jan de Vries, The problem of Loki 

FFC 110 

b. The Esthonian tale about the instrument of the thunder god 

In my wish to avail myself of the opportunity to con¬ 
sult the greatest amount of material attainable at this 
moment, I asked Dr. Loorits in Tartu to send me all the 
variants, that had been collected since the publication of 
Kreutzwald’s texts. He had the kindness to copy for me 
all the extant variants and I thought it advisable to insert 
them in the present paper. Meanwhile, however, I was 
informed, that he himself had prepared a publication of 
this material, accompanied by a critical treatment of the 
subject 1 ; therefore I shall refrain from doing this over again, 
because I am able to refer the reader to the investigation 
of Dr. Loorits. Still it will be necessary to mention some 
particulars, which may enable the reader, who has not the 
Esthonian material at hand, to follow the main lines of my 

The following variants are known at present: 

1 A. Kreutzwald’s text in »Inland», issue of 10. II. 1856 
Nr. 6 col. 89—90, originally collected by Johann Lagos 
from a peasant in Pernau. This Lagos, however, had 
the bad reputation of falsifying traditions by fanciful 
literary treatment. 

1 B. An other text, published by Kreutzwald in »Eesti 
rahwa Ennemuistsed jutud» (Helsinki 1866) p. 123— 
—126. This is a new and arbitrary adaptation of the 
former text. 

2. A third text of Kreutzwald, published in the same 
book; this is again an artificial contamination, where 
it is difficult to point out the real popular tradition. 

3. A short popular variant, collected in Kadrina 1898 
(E 35370); certainly under the influence of Kreutzwald. 

4. An other text from Kadrina 1899 (E 39389), showing 
the same character as 3. 

1 It appeared meanwhile in the Sitzungsberichte der Gelehrten 
Estnischen Gesellschaft 1930 under the title Das marchen vom 
gestohlenen Donnerinstrument bei den Esten (Tartu 1932). 

FFC 110 Chapter V, Loki as the companion of Thor 


5. A variant from Ambla 1895 (E 20718—22); the author 
of this text says himself that he added several words 
to ameliorate the story. 

6. Very short text from Koeru 1891 (H II, 39, 264) by 
a correspondent who is one of the greatest visionaries 
among the collaborators of Hurt. 

7. Very long and complicated story from Saaremaa 
(= Osel), published by M. J. Eisen in his »Eesti imede 
ilmast» (Tartu 1896); it shows the same shortcomings 
as the former texts: unreliability by personal additions 
and alterations. 

8. A text written down by a reliable correspondent in 
Kopu 1897 (H II, 59, 388—91), but from the mouth 
of a labourer, who came from Saaremaa. 

9. A variant from Tartu 1891 (H II, 44, 41—43), an 
apparently quite popular tradition. 

10. A short popular text from Karksi 1890 (H II, 23, 

11. Two texts, written down in Helme, A in 1910 (E 47396) 
by the father of the man, who furnished B in 1893 
(E 3238—9); the tradition seems to be reliable. 

12. A short variant from the same district 1893 (E 3046—7). 

13. A text of Wiedemann (1,125—6), probably from Voru- 
maa earlier than 1873. 

14. Legendary variant from Vastseliina 1889 (H II, 3, 

15. Very elaborate legend from the same district 1898 
(H II, 61, 287—296). 

16. Legend, called Ilja paiv from the same district 1898 
(H II, 61, 663-676). 

We see at once that the material is very unequal. We 
possess several genuine traditional texts, but the majority 
is, if not utterly worthless, at any rate of very questionable 
value, because the correspondents have treated the tale 
quite freely in order to give a more interesting appearance 
to it, or because they are under the strong influence of 
Kreutzwald’s inventions. It is to be observed that the 
reliable texts are all found in the South-Eastern parts of 
Esthland, with the only exception of 8, considered by 


Jan de Vries, The problem of Loki 

FFC 110 

Loorits as a genuine tradition and belonging to the island 
Saaremaa. The doubtful variants have been collected in 
two different areas, one in the North-East (3—6) and an 
other in the South-West, including Saaremaa (1, 2 and 7). 
This clear geographical distribution makes it easy to study 
this popular tale; of course we must begin with the authen- 
tical South-Eastern group. 

But here we must again make a distinction, for we can 
leave aside the variants 14—16, because they are properly 
speaking a legendary tale about Saint Ilja, the well-known 
successor of the thundergod in Slavic Europe. Then 9—13 
(especially 9—11) are apparently excellent examples of 
the real popular form. The tale runs ordinarily as follows: 
The devil steals the thunder instrument (called pikse riista, 
muristemise pill, muristamisasjad ; hence by no means a 
bagpipe as is often said) from the thundergod, in the vast 
majority of cases without any reasons or details. In 12 it 
is a young man, who steals it for him, but as he is called a 
fisherman, this is obviously a reminiscence of an other 
episode, that in its turn has been dropped. The thunderer 
who was asleep, is unable to find his instrument, when he 
wakes up; so he tries to get it back, often by taking service 
with a fisherman, while he assumes the appearance of a 
boy (9), or he goes fishing himself (10) and sometimes he 
is said simply to start on an expedition to find it (12—13). 
While he is fishing the devil takes the fishes from his net, 
but the thundergod sees this and succeeds in capturing 
the thief; the devil (or his son) says that he wants the fishes 
for a wedding-party and he is finally released after having 
invited the fisher and his servant to the feast (9—10). Most 
variants are very fragmentary in this part of the story, but 
they all agree in the description of the feast. Sometimes 
the thundergod arrives there by pure accident (12—13), 
or he goes with a companion without any apparent rea¬ 
son (ll). At the feast the devil is induced to fetch the 

FFC 110 Chapter Y, Loki as the companion of Thor 


instrument, which he keeps behind seven locks, but nobody 
is able to play on it with the exception of the fisherman’s 
servant. As soon as he puts it to his mouth, a thunder¬ 
storm arises, the devils are destroyed and the rain is again 
allowed to pour down on the earth. 

The legendary variants have preserved the different 
motives very well; they contain the theft of the instrument, 
the service with the fisherman, the capture of the devil in 
the net and the wedding-party. If we compare the other 
variants we find on the whole the same episodes, more or 
less changed and sometimes absent. On the other hand we 
may point out in some very few cases slight details which 
do not occur in the usual form. 

It is significant that these deviations are found in the 
introduction, the most vulnerable part of a popular tale. 
In the variants 2, 3, 8 and 12 the devil succeeds in stealing 
the instrument by the help of a servant, whom he raises 
to heaven by stretching out his neck. The thundergod lies 
asleep with his arm on the instrument; by the trick of the 
louse he is induced to make an movement and he therewith 
allows the thief to snatch it away. This is told in 2, 3, 5 
and 8 and belongs apparently to different regions of the 
country, but in texts that are either composed by Kreutz- 
wald or under the suspicion of having been submitted to 
his influence. 

There is still an other point of interest. The thunderer 
may act on his own behalf, or he is in a subordinate position 
with regard to an other and more powerful supernatural 
being. In the latter case he is obliged to confess the loss of 
his instrument; sometimes he actually does so, but some¬ 
times he dares not face the wrath of his master. This is of 
course the case in the legendary tales, where Ilja is the real 
hero; then he tells his misfortune to Taiva Esa (14) or it 
is said that he does not venture to go to Jummal (15). Loo- 
rits distinguishes them as two different redactions and this 


Jan de Vries, The problem of Loki 

FFC 110 

is methodically unimpeachable, but it is by no means sure, 
that they can be treated as two originally independant 
forms of the tale. On the contrary, the difference is too 
slight to be of much importance for such a far-reaching 
conclusion; they may be simply more recent ramifications 
of one original tale. It is, however, not clear at first sight, 
what the exact relation between them has been, for there 
are two equally plausible solutions: the Ilja-legend is the 
original form, where of course the saint must be sub¬ 
ordinated to God and then as a result of a later development 
the tale has been simplified by dropping the celestial being, 
that plays no r61e whatever in the tale. But we may also 
suppose that there has been a heathen tale about the thunder- 
god, in which he was not said to have been dependant upon 
a higher deity; then he was afterwards considered as a 
divinity subject to the heavengod, if he had not been de¬ 
throned by the influence of the Christian Ilja-legend. 

Dr. Loorits distinguishes three redactions. Besides 
A and B, respectively one where the devil acts on his own 
account (2, 3, 5, 8, 12) and the Ilja-legend with some con¬ 
nected variants (1, 6, 7, 9, 14—16), he mentions as a third 
redaction C the text 11, where we do not find a fisherman 
but a beggar. The thundergod, in the shape of a boy, meets 
him by pure accident and proposes that they should pay 
a visit to the christening feast of the devil’s mother. This 
is indeed a very unusual form of the tale. The difficulty 
is enhanced by the fact, that this text belongs to the same 
district as 12, a poor text, from which nearly all the details 
have disappeared, but where the fishing scene and the 
wedding-party are clearly mentioned. I do not think it 
worth while to ponder on this redaction; in my opinion the 
variants 11 A and 11 B are relics of a fragmentary and 
confused tradition; there are so many instances of the vivid 
imagination of Bsthonian story-tellers — and the corre- 

FFC 110 Chapter V, Loki as the companion of Thor 


spondent of 11 B was a schoolmaster! 1 — so we may 
safely assume, that the man who told this story, himself 
invented the beggar, because he had forgotten that it ought 
to be a fisherman. The christening party of the devil’s 
mother is likewise quite spurious; the correspondent of 
11 B added to it: »or some other feast». 

The motive of the louse is found in 2, 3, 5 and 8. Accord¬ 
ing to Loorits 3 and 5 are influenced by Kreutzwald, but 
8 must be treated as a genuine tradition. If this is true, 
this text should be particularly interesting, because it 
might give the solution for one of Kreutzwald’s sources. 
In this variant we are told that a musician Tiit (pilli Tiit ) 
steals the instrument at the devil’s request. Loorits sup¬ 
poses that this personage may have been the prototype of 
the paristaja poeg in Kreutzwald’s text 2. But this is by 
no means sure. The idea of a helper of the devil is found in 
more variants, as we have seen, f.i. in 12. Nowhere is he 
called a musician with the only exception of 8, where this 
gives the impression of a rather awkward innovation. 

But the popular character of the variant 8 seems to me 
on the whole rather doubtful. I wish to point out the follow¬ 
ing indications of literary influence: 

1. When the devil hears that the thundergod may possibly 
wake up, he says stammering: Kaak, kaak, kaak, kas 
paergus? (Will this happen presently?). And again, 
when Tiit asks for a louse, the devil answers: Mim, mim 
mitu tarvis on? (How many do you want?). The louse 
is described as so very large that Tiit lifts it up with 
both his hands! 

2. If the country Baruk, as Dr. Loorits suggests, has got 
its name from the biblical Baruch, this is an instance of 
untraditional elements. 

1 His father, who first had told him the story, afterwards com¬ 
municated it himself also to a folklorist; but of course he may have 
handed on the form, his son had given to it. 


Jan de Vries, The problem of Loki 

FFC 110 

3. The part of the tale, where the thunderer comes to the 
devil’s feast is quite out of joint. The devil fetches a 
man to play on his wonderful instrument. We find the 
same motive in 4, a variant that has only preserved the 
hare outlines of the story. It is obvious that our variant 
8 represents a partially defective tradition. 

4. The pill is closed behind seven locks. This is of course 
a detail from biblical tradition and may be due to literary 
influence. It is found in the texts 1, 2, 5, 8 and 12. The 
variants 1 and 2 are composed by Kreutzwald, 5 is in¬ 
fluenced by them, 12 is very corrupt. If the text 8 does 
agree in more cases with the Kreutzwald-redactions, it 
seems to me fairly certain, that it has also undergone 
this influence. 

5. The devil stretches out his neck to lift the thief up into 
heaven. This is found in the variants 2, 3, 8 and 12. As 
to the distribution of the texts, I may refer to the pre¬ 
ceding case. Since it is a very logical motive, that the 
devil must make some effort to reach as high as heaven, 
it may have been introduced by a redactor, who wished 
to furnish a fuller description of the events. 

6. The trick of the louse is found in the variants 2,3,5 and 8. 
Here the case is still more clear, because var. 12 does 
not even appear. Besides 8 we have only texts of Kreutz¬ 
wald or influenced by him. Certainly this motive does 
not properly belong to this story, for it forms part of 
quite a different folktale (Aarne-Thompson Nr. 560) 
and it has been introduced here to adorn the rather 
meagre story. Who has done it, Kreutzwald or popular 
tradition, represented exclusively by 8? The answer can 
not be difficult, since this is quite in accordance with 
Kreutzwald’s method of literary treatment. 

7. The thundergod shows himself in his real form as soon 
as he has taken hold of his instrument. This is only 
found in 2 and 8. In the other variants he keeps his dis¬ 
guise; sometimes, however, one of the devil ssuspects 
that he is more than his appearance shows, f.i. in 13 he 
says: this is pikse himself, for I have recognised him by 
his red eyes (cf. also 7, 9,11 and 16). The circumstance, 
that in this respect 8 corresponds with 2 and with no 
other variant, almost proves that 8 depends on Kreutz¬ 
wald’s text. 

FFC 110 Chapter V, Loki as the companion of Thor 


These are my reasons, why I arrive at a conclusion 
different from that of Loorits. The redaction of Saaremaa 
is also an artificial text, going back to the famous collections 
of Kreutzwald. It was not necessary for the informant to 
have read a literary form, because there had been for several 
years such a continuous influence upon popular tradition, 
that one could repeat a text of Kreutzwald without being 
conscious of it. Then there is no reason to believe in the 
existence of two independant traditions in Saaremaa (7 and 
8) which Dr. Loorits considers of great importance. More¬ 
over, the variant 8 was taken down from the mouth of a 
labourer in Kopu, and although this man was born in Saare¬ 
maa, this does not imply that his text belongs to the tradi¬ 
tion of this island; he may have picked it up elsewhere as 

Loorits arrives at the conclusion that the A-redaction 
may have a Scandinavian origin, belonging properly to the 
viking-traditions. The only valid argument seems to be 
the motive of the biting louse, that reminds one of the myth 
about the Brisingamen. The conditions in both tales are 
quite different; there can be no organic connection what¬ 
ever. If, as I have tried to prove, the variant 8 is dependant 
upon a text of Kreutzwald, then we may explain the co¬ 
incidence between his tale and the Scandinavian myth 
simply by the assumption, that he himself has imitated 
this interesting detail which he found in a popular tale. 
When the A-redaction has, as we have shown, such a very 
doubtful character, it is rather unsafe to make use of it 
for a hypothetical construction, according to which viking- 
traditions should have influenced this tale. 

The B-redaction has, on the contrary, a Russian origin. 
The legendary variants about IIja can only be derived from 
a Russian source, although such a tale has not yet been 
found in Russia. Here we find all the most interesting 
motives of the tale: the fishing scene and the wedding feast. 


Jan de Vries, The problem of Loki 

FFC 110 

If there has ever been an indigenous Esthonian »myth» 
about the theft of the thunder-instrument, it is likely that 
it has had a very simple structure, and that it has developed 
into the actual form by a corresponding legend about Hja. 
Penetrating farther into Esthland, the Christian names 
were dropped and so the tale obtained the appearance of a 
real heathen myth. But, although Ilja very easily could be 
given up, the devil could not be ousted and he acts in all 
the known variants as the malevolent opponent of the 

In which way Kreutzwald collected his material seems 
to be as yet an open question. Loorits tries to give a solu¬ 
tion, but in a very hesitating way. I do not think it prob¬ 
able that he got any information from Saaremaa, as this 
supposition is only founded on the alleged originality of 
the variant 8. The best and most complete redactions are 
found in the South-Eastern parts of the country, in Voru- 
maa and Setumaa; why should he not have heard an example 
of this tradition? It is even very likely that he knew the 
tale in the form of a legend, and that he has dropped the 
name Hja and given to it a more heathen form. Is it indeed 
not fairly certain, that the Lij on-mystification in the text 
of Lagos ultimately goes back to the figure of Ilja 1 ? In 
Kreutzwald’s text 1 we read at the end, that the thunder- 
god says to the fisherman Lij on, that as a token of gratitude 
he will never refuse any wish expressed by him. And ever 
since Lijon has remained a mediator between the gods and 
mankind. The same is related in the second text of Kreutz¬ 
wald (I B), but in the variant 2 the fisherman has no parti¬ 
cular name and it is the thundergod himself (paristaja poeg ) 
who becomes the servant of the heavengod ( Koutaat ). This 
agrees very well with the legendary forms, where Ilja is 
also the servant of God; in variant 15 he says to the old 

1 Cf. the interesting discussion in Loorits’ paper on p. 102 ff. 

FFC 110 Chapter V, Loki as the companion of Thor 


fisherman: »Dear master, remember this: if I drive through 
the clouds in my golden carriage and you hear the cracking 
of my whip and the sound of my pill , then all your nets 
will be filled with fishes. This will be your reward.» Both 
motives, the gratitude of the thundergod and his obsequious¬ 
ness to the Lord, are found also in the adaptations of Kreutz- 
wald. It is a queer irony of fate, that Olrik and Celander 
have had such an acute controversy on account of a person¬ 
age, who proves to be the result of a quite arbitrary adapta¬ 
tion of Kreutzwald. 

Loorits arrives at the conclusion that Kreutzwald is 
still of some importance in the characteristic features of 
the tale for the student of mythology. To judge from the 
example, presented by the tale of the thunder-instrument, 
I am not so sure about this reliability for the investigation 
of popular traditions; if we have no real genuine variants 
to correct his artificial adornments, we had better refrain 
from this troubled source. At any rate the famous Estho- 
nian »myth» of the theft of the thunder-instrument is in 
its present form originally a Christian legend and there is 
no connection whatever with any viking-tradition. The 
form, Kreutzwald has given to it, is nothing better than 
a mystification, to use the severe but justified term of Dr. 
Loorits. And such a worthless falsification has been treated 
by several scholars as a reliable text, learned digressions 
have been written about it and the attempt has been made 
to solve difficult questions of Scandinavian mythology by 
means of such a forgery. Here again we percieve how danger¬ 
ous it is to make use of popular material in matters of reli¬ 
gion, if we have only the faintest ideas about the character 
and the provenience of these traditions. 

c. The Finnish tale about the instrument of the thundergod 

Besides the Esthonian tale there exists also a Finnish 
story about the way in which the thundergod got his instru- 


Jan de Vries, The problem of Loki 

FFC 110 

ment back. As the Finnish variants show a rather great 
difference from the tale, as it is current in Esthonia, we 
must give a short survey of their contents. Prof. Kaarle 
Krohn has been so kind as to send me the most interesting 
variants, that could help me in a study of this tradition; 
of the 37 texts tabulated in Aarne’s list 1 2 under Nr. 1148 
there are 10 variants that belong to this type, viz. the 
numbers 2 (from North Tavastland), 5 and 6 (from North 
Savolax), 12 (from East-Karelia), 14,15 and 17 (from North- 
Karelia), 26, 27 and 29 (from Middle Osterbotten). The va¬ 
riants 15, 17, 26 and 29 are more or less fragmentary, whe¬ 
reas 6 is told in a rather equivocal literary style. The con¬ 
tents, however, agree very well and it is not difficult to 
establish the original form of the tale. 

During his sleep the thundergod (Ukkonen) or God 
(Jumala ) is robbed by the devil. His instruments are called 
pelivarkit (26) or mill-stones (6: ne kivet joilla ukko jurisoo). 
In 14 the devil brings the instrument of thunder (jylistimet) 
to a rock. In disguise the god goes to the devil and offers 
to become his servant. Then they embark to fish together; 
•the servant proves to be exceptionally strong. The devil is 
induced to boast of his having robbed the thundergod of 
his instrument and to fetch them (in some variants they are 
locked up in a room with many bolts). The devil, however, 
can not play on them; hence the strong servant may try his 
forces. He begins to play softly, but gradually with in¬ 
creasing violence and at last the devil’s hall bursts asunder, 
his children are smashed to pieces and he himself dies, is 
thrown into the air or swoons. 

In his discussion of the Lappish, Finnish and Esthonian 
tales Kaarle Krohn 2 points out, as Axel Olrik had done 

1 Finnische Marchenvarianten, FFCommunications Nr. 5. 

2 Ubersicht it bar einige Resultate der Marchenforsehung, FFCom¬ 
munications Nr. 96, p. 124 sqq. 

FFC 110 Chapter V, Loki as the companion of Thor 113 

before, several seemingly strong resemblances with the 
Eddie poems jrymskviSa and HymiskviSa. It may, how¬ 
ever, be advisable to treat the popular traditions of Fin¬ 
land and Esthonia separately, because they show a few 
remarkable differences. The frame of the tale is obviously 
exactly the same: the instrument is stolen, the thundergod 
disguises himself as a servant, and likewise the denouement 
is the same. Still there are some interesting details in which 
they deviate. In the Finnish tale, which on the whole is 
much simpler than the Esthonian, the thunderer becomes 
the servant of the devil himself; hence the fishing-scene 
is told in an altogether different way. Now, according to 
Krohn, the more complicated form of the Esthonian tale 
is a later development of a shorter story, identical with 
the Finnish; the robbing of the fishes out of the net may 
be a repetition of the theft of the thundergod’s instrument 
and the being entangled in the net is a motive borrowed 
from other folk-tales. This is, indeed, quite possible; at 
any rate there is too great a similitude between the Finnish 
and the Esthonian tales to consider them as parallel, origi¬ 
nally unrelated traditions. Now we have arrived at the 
conclusion that the B-group of the Esthonian variants is 
most probably of Russian provenience. Dr. Loorits supposes 
the A-group to belong to viking-traditions, but the only 
motive, pointing in this direction, is the episode of the 
biting louse — and of this motive there is no trace what¬ 
ever in the Finnish tale. Hence we are induced to explain 
the curious relation between the Finnish and Esthonian 
tales by one of the two following suppositions: either the 
Esthonian tale is the original form and during its wandering 
td the Finnish people it took a different form by dropping 
several episodes and readjusting the remaining ones, or 
else the Finnish tale represents the original and in Esthonia 
it was combined with a Russian legend about Ilja, 

I, myself, am more inclined to the former supposition, 



Jan de Vries. The problem of Loki FFC 110 

but I avow readily that I am unable to give any decisive 
proofs for my opinion. 1 Hence 1 wish for the moment to 
accept Krohn’s view as to the originality of the Finnish 
tale and to consider on this basis the possible relations be¬ 
tween this story and the Eddie poems. The objections that 
may be raised against this assumption are the same that 
oppose themselves against Olrik's hypothesis about the 
relation between these Old-Norse poems and the Esthonian 
tale. I shall discuss them in detail in the following para¬ 
graph. Here this may be stated: the fishing-scene has indeed 
some similarities with the famous episode of the Hymis- 
kvida. but only in this respect that in both stories a god 
proves his superior strength while rowing with a giant in 
a fishing-boat. This would be conclusive if only the Eddie 
poem related a kind of hamarsheimt, but the Hymiskvida 
tells about the fishing of the world-serpent. Then we are 
not justified to construct a different Scandinavian myth 
in which Thor entered into the possession of his instrument 
by means of the trick, that he took service with the giant, 
who robbed him of his hammer. And if we did, we should 
only arrive at a supposed tale, in which there is no imagi¬ 
nable role for a personage like Loki. In the Hymiskvida. 
as well as in the Finnish tale the god is not accompanied 
by a servant, but he is himself a servant. 

The identity of the Eddie poem and the Finnish talr 
consists exclusively in the rowing contest between the god 
and the giant. This may have been originally a myth, but 
then it looks deceptivily like a common folk-tale. It is 
quite in accordance with the well-known group of stories, 
in which a strong boy enters the service of a giant and by 
his cunning proves stronger than his master. 2 The only 

1 The same may be said, of course, with regard to the opposite 
assumption; here decisive arguments are wanting likewise. 

2 Aarne’s number 1087 is a rowing-contest, but the boy wins bv 
damaging the giant’s boat. This motive is only found in Finland. 

FFC 110 Chapter Y, Loki as the companion of Thor 


difference is, that the thundergod indeed has more strength 
than the giant and that he makes him tremble before his 
unexpected force. Moreover among a people of fishermen 
such a tale could easily be imagined. If the fishing of the 
world-serpent should be a Scandinavian adaptation of a 
Christian legend about Leviathan, then we are likewise 
impelled to explain the rowing-contest in the HymiskviSa 
as a new motive introduced into the Old-Norse rendering 
of the legend. Is there any plausible reason to suppose 
that it had been previously part of a story about a hamars- 
heimt and is it not much simpler to regard it as a motive 
current in folk-tales? Since it is moreover a motive that 
belongs to the kind, Wesselski very aptly calls »(femein- 
sehaftsmotive», there seems to me no necessity whatever 
for the far-fetched supposition of a lost myth about a Scandi¬ 
navian hamarsheimt. For the study of Old-Norse mythology 
about the thundergod and his servant the Finnish tale 
seems to me of no value, indeed of less value than the Estho- 
nian tradition, to the discussion of which we will now return. 

d. Criticism of Olrik's hypothesis 

It will now be necessary to take up the problem in its 
entire compass. The questions, that require to be answered, 
are the following: Have we any indications for the supposi¬ 
tion that Loki and tqalfi belong to different areas of Scandi¬ 
navian mythology? Is £>jalfi originally a Swedish form of 
the thundergod’s companion? Is Loki on the contrary 
confined to another area and may we consider the pikkeri- 
pois of the Esthonian myth as an early and more original 
form of Loki? 

I agree with Axel Olrik in this respect, that a little 
companion of the thundergod--a not uncommon notion 
in primitive belief — has been known in Old-Norse mythical 
conceptions of this phenomenon. Perhaps he is the religious 

FFC no 

116 Jan de Vries, The problem of Loki 

expression for the faint rumbling before the actual peal of 
thunder, although this explanation is by no means sure. 
The small figure at the side of the thundergod, as it is 
depicted on the magic drums of the Laps, is sufficiant evi¬ 
dence for this conception in Lappish and we may infer that 
it has also been known in Swedish belief. But the Swedish 
folk-tale about »lyfte-stenen» in the Linneryd-district can 
not be taken in its full extent as a genuine heathen tradition. 

Thor meets a giant on one of his journeys, who wishes 
to fight him. But the god mocks him and says that he can 
not even lift the small pebble that lies on the top of a huge 
stone standing near by. Indeed the giant is not able to 
do it. Then Thor tells his small servant to try to do it and 
he does it without any difficulty. The giant gets into a 
fury, but is smashed to pieces by the powerful hammer of 
the thundergod. 

It is evident that this tale is invented to account for the 
curious stone in Linne'rydsby. This stone was a kind of 
oracle, even in the 18 th century when Rudbeck told about 
it in his »Sm&lands Antiqviteter». The churchgoers were 
wont to try their strength by lifting the small stone and 
curiously enough, often an apparently feeble man succeeded 
in this effort, while a stout and sturdy person could not 
even move it. In itself this is a popular belief 1 , that has 
nothing whatever to do with the person of Thor. But he 
might be brought into some connection with it and then 
it lay near at hand to make the lifting of the stone the 
object of a contest between him and a troll. It is, however, 
not a superiority of muscular strength that gives the deci¬ 
sion in this case; the small stone has no excessive weight 

1 Cf. H. F. Feilberg, Bidrag til en Ordbog over Jyske Almuesmal 
II, p. 551—2; Von Unwerth, Untersuchungen iiber Tolenkult und 
Odinnverkehrung bei Nordgermanen und Lappen, p. 12 ii.\Eric Elg- 
qoist, Nagot om Lyftestenar, Nordiskt Folkminne, Studier tilliignade 
C. W. von Sydow p. 71—84. 

FFC 110 Chapter Y, Loki as the companion of Thor 117 

but it has mythical properties by which some people are 
allowed, to lift it, others not. Thor consecrates the stone 
with his hammer and that is the reason why the giant is 
unable to lift it. In a Christian legend Thor might be re¬ 
placed by a saint and then the token of the cross would 
have had the same effect as Thor’s hammer. 1 Hence it is 
not the force that decides, but exclusively the fact if the 
magical potency is propitious or not. The small servant 
of Thor illustrates this in a very interesting way. 

It may be asked whether this popular tradition does 
not show any traces of its being influenced by the Eddie 
literature. Olrik denies it flatly. The author, however, 
who preserved this curious instance of folk-lore for us, 
knows the Edda very well and he himself makes an allusion 
to the Eddie lyjalfi. 2 Hence we do not know in what degree 
he has fashioned the popular tradition into its present form. 
We are not quite sure, if the informant Nils Knutsson has 
not himself introduced Thor into the story, perhaps because 
his interlocutor had asked him, if he knew anything about 
this god. 3 But since we have no indication whatever in one 
direction or another, let us accept this tale as genuine. 

What then does it tell us about the heathen conceptions 
of Thor and his servant? Olrik points out the great differ¬ 
ence between the character of Eddie mythology and popular 
traditions and he arrives at the conclusion that in this 
respect the myth from Smaland represents an earlier degree 
of civilisation (»et addre kulturtrin») than almost the entire 

1 Or the tale may have been originally a Christian legend and 
afterwards have been remodelled by antiquarians into a »heathen 
myth». There is a similar Mohammedan tale about Moses and a 
certain Schoeib; the latter is unable to lift a slick from the ground, 
but Moses does it without any effort (cf. A. Wunsche, Die Sagen vom 
Levensbaum und Lebenswasser p, 45). 

2 »Och hade en drang, och lar det vara den Edda kallar Tjolfeo. 

3 Cf. my observation in FFComm. Nr. 94 p. 5. 

118 Jan de Vries, The problem of Loki FFC 110 

Eddie poetry with its aesthetic proclivity to characterisa¬ 
tion. That is the reason why we may infer that the simpler 
conception of the popular tale represents a more primitive 
form of the religious representations than the literary pro¬ 
ducts in the Edda. 

This is again a curious example of the overestimation 
of popular traditions. Of course a tale, current among 
peasant-people and obviously having lived many centuries 
in the memory of a Christian community, does not show 
the same character as the highly artistic products of Ice¬ 
landic literature. But it does not follow that they should 
be older nor that they are more genuine. On the contrary, 
the adaptation of a literary product to the mind of un¬ 
cultivated people always brings about a transformation of 
the original tale and a reduction to simpler forms. If the 
tale of the »lifting-stone» presents to us the servant of the 
thundergod as a mere companion, without attributing to 
him any particular role, this need not be explained as a 
proof of its high antiquity. In the economy of the tale there 
was only a very restricted place for the servant; consequently 
his character was accommodated to the more primitive 
setting of a local tradition. The conception of the servant 
in the Swedish folk-tale does not imply that originally the 
mythical value of j-jalfi has been quite the same; it only 
shows us what has been the result of the simplifying develop¬ 
ment in the course of popular tradition. 1 

The »myt,h» of the lifting-stone is not of such exceptional 
importance as Olrik supposed it to be. We learn from it, 
what we already knew: Thor has a little companion on his 
journeys. But he may have been the same intelligent person 
as in the myth of Hrungnir. The folk-tale, however, only 

1 The same may be said with regard to Olrik's treatment of the 
Lappish tale of the bound Termes (Danske Studier 1906, p. 65 ff.); 
it may be as well a simplification of a more elaborate myth, or this 
myth on an older cultural level, as Olrik takes it. 


FFC, 110 Chapter V Loki as the companion of Tiior 

required him as a foil to both Thor and the iriant; hence 
only his small stature is mentioned. 

In Sweden the conception of the thundergod’s servant 
seems to have been known. Was it I>jalfi? Olrik tries to 
prove it, although in a very unsatisfactory way. Obviously 
it is not enough to point out that the Lappish belief and 
the Swedish folk-tale both suppose it to have been known 
among the Swedes; for why should it have been more at 
home here than everywhere else in Scandinavian territory':' 
As the sole support for his view he mentions the fact, that 
bjalfi is a rather common name in Sweden, whereas it 
seems to have been unknown in Norway. Lundgren whose 
authority Olrik invokes, speaks about »the not so very 
rarely occurring» name and asks hesitatingly, if this name 
has been given on account of the mythological pjalfi. 1 
We do not know why such a name has had particular favour 
in one paid of Scandinavia, why not in another; but it may 
lie quite independant from the original mythical meaning 
of-the word. 2 The founder of Clot lands population about 
whom the Uutasaga tells a curious tale, is also called t>jel- 
var; the connection with the thundergod’s servant is by 
no means clear. Furthermore the name is not quite un¬ 
known in Norway. E. H. Lind quotes 3 4 a local name pialfu- 
hrllir as possible proof for its being used as a personal name. 
The surname j orkell J ialfi, if this is the correct form and 
not bialft may point in the same direction. One might 
suggest that the reason why the name was only very seldom 

1 Spar af hednisk tro och kult i fornsvenska personnamn p. 4o. 

2 As some of the books quoted by Lundgren are not accessible to 
me. 1 am unable to verify, if the examples belong to a particular 
district of Sweden or not. 

3 Norsk-isliindska Dopnamn och fingerade namn fran Medeltiden 
col. 1123. 

4 E. II. Lind, Xorsk-islandska Personbinamn fran Medeltiden col. 


Jan de Vries, The problem of Loki 

FFC lio 

used in Norway was because it denoted a superhuman being 
and that it on this account could not be appropriately used 
as a personal name. 1 On the other hand its more frequent 
occurrence in Sweden might indicate that here the mytho¬ 
logical personality was scarcely known. But from what 
has been said it will be clear that the facts are scanty and 
admit of different explanations. 

I arrive at the conclusion that the reasons, adduced by 
Olrik for a special connection of J>jalfi with Swedish belief, 
are unsatisfactory; on the other hand it is evident from the 
literary traditions that he has been in Norway and in Ice¬ 
land more the companion of Thor than Loki, who only 
occasionally plays this role. 

What is the value of the Esthonian myth? Olrik finds 
a strong resemblance between it and the Icelandic tradi¬ 
tions about Loki, but he was not the first to observe this. 
As early as 1858 Jacob Grimm after reading the first prin¬ 
ted text in »Das Inland» noticed the identity with the Jryms- 
kvi&a and the HymiskviSa, as is proved by a communica¬ 
tion of Schiefner in the same periodical. 2 Olrik makes use 
of two versions published by Kreutzwald in his collection 
of Esthonian folk-tales and translated by H. Jannsen. The 
first text (Nr. 4 of the second part of this translation) is 
the same as Nr. 2 in the list of the variants, mentioned 
above, and the second text (Nr. 10 of the first part of the 
same translation) is Nr. I B of the same list. 

Olrik points out the following similarities. 3 The thunder- 
instrument is stolen and recovered, as is the hammer in the 
t>rymskvi&a. The thundergod acts as a servant of a fisher¬ 
man and catches the devil in his net, cf. the HvmiskviOn 
where Thor draws the Midgardserpent from the bottom of the 

1 Names of gods are common as part of a composition, but not 
without a second element. 

2 Ibidem 1858 Nr. 39 col. 629. 

3 Danske Sludier 1905 p. 142. 


FFC mi Chapter V, Loki as the companion of Thor 

ocean. Then the son of the thundergod obtains his release 
from captivity by promising the devil to steal the instru¬ 
ment; Olrik compares the myth of Ibunn. The theft suc¬ 
ceeds by means of the trick of a flee, as we found it in the 
myth of the Brisingamen. Finally the pikkeri pois plays 
the same rote as the Loki of the prymskviSa. 

In mv opinion this comparison shows three serious 

1. If we find the various motives of one tale elsewhere 
in four different traditions, the value of these similitudes 
is seriously weakened. They may have been after all several 
loose motives, common to the belief or popular traditions 
of both peoples but combined into one single tale only in 
Esthonian folk-lore. This is perfectly clear with regard to 
the motive of the biting louse. If we do not accept this 
explanation, there are only two possibilities: a the Estho¬ 
nian folk-tale has borrowed these motives from Scandina¬ 
vian tradition as parts of the Loki-cycle and united them 
into one single myth. This is of course very improbable, 
as we do not see how the Esthonians could have become 
acquainted with them. b. The Esthonian folk-tale pre¬ 
supposes that among this people traditions about a servant 
of the thundergod have been current, going back to pre¬ 
historic borrowings from a Scandinavian people. This seems 
to be the view of Olrik, but it is supported by no proof 

2. The similarities are on the whole very superficial. 
Catching the devil in a net, where he steals the fishes is not 
the same as drawing up the Midgardsserpent by means of 
a fishingline. The hammer is not the same as the pill. The 
marriage feast of the devil has quite a different meaning 
from that in the | rymskviSa. The trick of the biting louse 
is a simple motive of folk-tales. Only by reducing these 
motives to their most abstract form can we get a likeness, 
that is at first sight not altogether improbable. 

1” Jan de Vries. The problem of Loki FFC 110 

d. Tile texts used by (ilrik arc unreliable, as they have 
been composed by Kreutzwald in a most arbitrary way. 
It is certainly excusable in the case of Olrik, who did not 
know more variants of this tale, but it shows at the same 
time - as already said above — how exceedingly dangerous 
it is, to draw farreaching conclusions from slight material, 
that moreover is not absolutely reliable. 

Now we have a larger store of variants, which are more¬ 
over partly really popular; so we may ask again if the 
Esthonian myth proves anything for Loki as the companion 
of the thundergod. A servant of this god is to begin with 
quite unknown in Esthonian folk-lore. Here we find two 
different figures, on the one hand Kou or Koue, on the other 
hand Pikne or Piksepoiss. The first one is the real thunder- 
god, often hardly distinguishable from the heavengod 
Vanaisa or Vanataat. The relation between him and Pikne 
is quite obscure; now-a-days Kou is represented as an old 
man, whereas Pikne can be a boy as well as a grown-up 
man. In var. 2 (Olrik’s A-text) the thundergod is named 
Kou-taat, Yana taat, Miiristaja taat and Pikker, a queer 
medley of names, showing the arbitrary character of Kreutz- 
wald’s compilation. But in a prayer to Kou, written down 
in the village of Yonnu, the god is addressed as Tuli-Pikker 
and as Kange-Kou. 1 If Eisen says that originally Kou has 
been the thunderer and Pikker 2 the god of lightning, I 
do not know on what fact he bases this opinion, but 
when he continues to say that Pikker also sounds the 
trumpet 3 , I do not see why he should have been different 
from Kou. 

For the sun of the thundergod we have only evidence 
from the Southern Esthonians, who say that piksepoiss 

1 (If. ./. M. Eisen , Estnische Mythologie p. 161. 

2 This is according to Loorits p. 97 an untraditional form for 

3 Ibidem, p. 162. 

FFC. 110 Chapter V, Loki as the companion of Thor 12!? 

sounds the pdugahuse star . 1 Eisen, who docs not know any 
explanation for this boy. supposes that he is the same as 
Pikne himself, because he often changes himself into a 
young man. 2 This seems quite acceptable. In var. 9 the 
thunderer is called Pikne. but when lie acts as the foster- 
son of a poor man. his name is Pikse poisihr. Hence the 
thundergod in his disguise as servant of a fisherman may 
be thought of as a small boy. 

The son of the thundergod does not belong to the Esto¬ 
nian myth. 3 In the original form of the tale the thundergod 
is robbed of his instrument, then he goes to seek it and takes 
service with a fisherman. Here he succeeds in catching the 
devil who steals fishes for the wedding-feast and Satan is 
only released after having invited him to the feast. In this 
part of the tale there is no place for the son of the thunderer; 
he only appears in the introductory part of some texts. But 
here more recent additions are only to be expected and in 
fact in most variants, particularly in those that are popu¬ 
lar, the devil himself steals the thunderinstrument. 

From these Esthonian tales we can not expect to get 
any light for the elucidation of the Old-Korse myths about 
Loki. They show in their original form only very slight 
similarities to the Eddie poems about Thor’s journeys 
to Hymir and j»rymr, in fact they go back to quite a differ¬ 
ent conception of the thundergod. The hammer in Scandi¬ 
navia is irreconcilible with the pill of Pikne or the j grist i- 
mei of Ukkonen. If the figure of the god himself is quite 
different, how could we expect that there would be con¬ 
formity as to his rather unimportant servant? In fact there 
is none, for the god Pikne changes himself into a boy, to 
get unnoticed into the devil’s house; it is consequently the 

1 Cf. in var. 13: porngrihiisse sa/v. 

2 Ibidem, p. 164. 

3 This is. of course, confirmed by our analysis of the Finnish 
folk-tale, where he is nowhere mentioned. 

124 Jan de Vries, The problem of Loki FFC 110 

pretended servant of the fisherman who succeeds in sound¬ 
ing the famous pill. 1 I can not detect any real likeness to 
the Scandinavian traditions, where Loki in some cases is 
the companion of the thundergod, but if he has the wit to 
escape from the difficulties, Thor has after all the strength 
necessary to annihilate, the trolls. On nearer examination 
all the advanced similarities prove to be unimportant or 
even simply fade into nothing. 

1 This is the case both in the Finnish and the Esthonian tale and 
must belong consequently to their original form. 


1. Criticism of the literary sources 

In the present chapter we have to face a problem of 
Scandinavian mythology that is particularly difficult owing 
to the scantiness and incompleteness of our sources. More¬ 
over there are two different myths that have nothing in 
common but for the names of Loki and of Freyja’s jewel. 
As one tradition is fragmentary, the other one unreliable, 
they are discouraging material for the scholar, who wants 
to arrive at a definite conclusion about the real meaning 
of this myth. 

The first tradition tells about a fight between Heimdall 
and Loki on account of the Brisingamen. Properly speaking 
we possess only some allusions scattered over several chap¬ 
ters of the Snorra Edda. The most extensive information 
is furnished by ch. 8 of the Skaldskaparmal. Here the 
author gives fur Heimdall a series of kennings: Hvita-as .s, 
Loka-dolgr and mensceldr Freyju. As an explanation for the 
latter circumlocutions he adds the following sentences: 
»Hann er ok tilsa kir Vagaskers ok Singasteins; l a deilbi 
hann vid Loka um Brisingamen; hann heitir ok Yindhler. 
Ulfr Uggason kva5 i Husdrapa langa stund eptir peiri 
frasogn. ok er Jess Jiar getit, at Jieir varu i sela-likjum.» 
Again in ch. in, where he speaks about Loki, he mentions 
among the kennings pjofr Brisingamens and J>rcelu-d6lgr 
Heimdalar; but adds moreover one stanza of the Hus¬ 


Jan de Vries, The problem of Loki 

FFC 110 

This stanza is our most valuable and practically our 
only source. Since, according to the Laxdtelasaga, this 
poem was written on the occasion of the building of Olafr 
pai’s famous hall, we know exactly at which date it has 
been composed: in 983, at a time when the Christian religion 
had not yet been officially adopted in Iceland, but foreign 
influences were rapidly submerging the old heathen tradi¬ 
tions. Unfortunately this stanza is very obscure; it runs 
in this way: 1 

RaSgegninn breg&r ragna 
rein at Singasteini 
fragr vi5 firna sloegjan 
Farbauta mog vari; 
moQoflugr rtebr moedra 
.mogr hafnyra fogru, 
kynnik 65, ok unnar 
atta, mgerdar pottum. 

We are obliged to enter here into the details of a philo¬ 
logical interpretation. The subject of the first half is, accord¬ 
ing to the interpretation, advocated by F. Jonsson: Vod- 
gegninn, frcegr ragna rein- / -vari, that is >>the ingenious, 
famous protector of the way of the gods». Here, however, 
there are two difficulties to be noted: 1 . rein has not the 
meaning of »way, road», but it denotes originally the un¬ 
tilled strip of land between two fields. It is not uncommon 
in skaldic kennings 2 3 , but then it always means »land», never 
»road». Hence it is more likely that Heimdall is called »the 
defender of the borderland of the gods». 2. The awful 
tmesis rein- / -vari is suspicious; E. A. Kock, who tries to 
do away with such artificial stylistic curiosities proposes * 
reinar. Then, however, the line counts seven syllables. 

1 Skjaldedigtning IB p. 128 but with a few emendations. 

2 Lexicon poeticum 2 p. 462 s. v. 

3 Notationes Norrcenae § 1952. 

FFC 110 Chapter Yl, Loki and (he Brisingamen 127 

F. Jonsson reads the half stanza »(Heimdallr) bregdr 
vid (Loka) at Singasteini» and translates »Heimdall goes 
together with Loki to Singasteinn». The rendering of bregdr 
at c. dat. as »to go to» is quite improbable; Jonsson says 1 
very prudently: this seems to be tin- most natural explana¬ 
tion, although bregda is nowhere used in tins meaning». 
E. .1. Kock 2 call this translation also rather impossible, 
but does not give a better one in stead of it. We might 
expect, however, that bregdr c. dat. of an object means as 
usually »to take suddenly away». Then the construction 
of the sentence must be the following bregdr Singasteini. 
Hence Singasteinn is not the name of a reef in the sea, but 
it denotes the jewel itself. Snorri has known the reading 
rein at Singasteini and he has.come to the conclusion that 
the proper name meant a certain place, obviously the same 
as Vdgasker. found in another stanza, which he unfor¬ 
tunately has not preserved for us. The better and older 
reading may have been reinar Singasteini an interesting 
confirmation of Kock’s suggestion to read reinar in stead 
of rein at. 

Heimdall takes the jewel away rid firna shrgjan Fdr- 
bauta mag. Vid c. ace. is not used to denote together with» 
but it means »in opposition to. against». Hence the mean¬ 
ing of this part of the stanza is: »Heimdall took the jewel 
away from Loki». 3 4 

For the second part of the stanza we must follow the 
interpretation, proposed by E. A. Kock*: kgnnik od nuerdar 
poll am, i.e »I show poetical genius in the different parts 

1 Lexicon poeticum 2 p. 62. 

2 Notationes Xorrcenae § 420. 

3 Another possible translation would be: »he violently struck the 
stone against Loki» (cf. the expression bregda e-u vid e-n)\ but this 
is rather difficult to bring into connection with the remainder of 
the stanza. 

4 Xotationes Xomenae § 1800 . 

128 Jan de Vries, The problem of Loki FFC lie 

of my poem». The rest consists of the words: modoflugr 
mogr dtta ok einnar ma’dra (= Heimdallr) rcedr fggru haf- 
nyra , which of course can only mean »he was in possession 
of the beautiful jewel» evidently as the result of his taking 
it away from Loki. 1 

This is all we know about this myth. Is it the beginning 
or the end of the story? For the tale has had a tolerable 
length; Snorri even says: Flfr Uggason kvad % Husdrapa 
lang a stund eptir peiri frasogu. Still his information 
is despairingly scanty. We are able to detect the following 

1. Loki took away the jewel; hence the kenning Jjjofr Bri- 


2. Heimdall made an effort to get it back; cf. the kenning 

uienscekir Freyju. 

3. Loki had a fight with Heimdall, apparently in the sea. 

4. They were both transformed into seals. 

Perhaps we may piece these scraps of information to¬ 
gether in this way: Loki had stolen the jewel, but Heimdall 
persecuted him and succeeded after a fight in the shape of 
a seal in recovering it. Then the stanza which has been 
preserved is the end of the story. But this is by no means 
certain, because we have no proof, that the kenning fijofr 
Brisingamens for Loki is in connection with this myth; it 
is even more probably that it refers to a myth, preserved 
in the Soria Jattr, to which we will turn presently. If we 
must leave out this kenning in the reconstruction of the 
myth of the Husdrapa, the circumlocution menscekir Freyju 
gains a much greater importance: it is Heimdall who made 
an effort to come into possession of the jewel and he did 
so by fighting Loki. 

But we may perhaps even go farther. The jewel is called 
hafnyra »kidney of the sea». This evidently denotes a kind 

1 Ibidem § 1052. 

FFC 110 Chapter VI, Loki and the Brisingamen 


of precious stone to be found in the sea; that is why the 
two antagonists have to fight for it in the shape of seals. 
It is a piece of amber, used as an ornament since the most 
remote period of prehistoric times. Vdgasker is a name for 
it, a clear kenning meaning »the rock of the billows». An 
other name seems to be Singasteinn . 1 

If we call this jewel, that came into the possession of 
Freyja — as is commonly done — Brisingamen , we are only 
combining information from other sources with this stanza 
of the Husdrapa; here, however, we do not find any allusion 
whatever to it. It is the authority of Snorri, which leads 
us to this statement, for he says in ch. 8 of Skaldskaparmal: 
(Heimdallr) er ok tilscekir Vagaskers ok Singasteins: pa 2 
deildi harm rid Loka um Brisingamen. As it is not safe to 
rely exclusively on the statement of Snorri and being on 
our guard against rash conclusions, we will return to the 
discussion of this question after having studied the second 
myth of Loki and the jewel. 

It is preserved in the Soria' attr, a rather recent tradi¬ 
tion inserted in the compilation of the Flateyjarbok. The 
main subject of the ' attr is the story of the famous Hja5- 
ningavig, that is said to have been brought to an end by 
I varr ljomi, a warrior of king Olaf Tryggvason. 3 An introduc- 

1 I am of opinion that these words can not be explained as geo¬ 
graphical name'.; it would be surprising if the locality, where the 
fight took place, had two names. Pipping' s interpretation of the 
names as poetical circumlocutions for the localities in Uppland Singo 
and Vaxon (cf. his Eddastudier III in Studier i Nordisk Filologi 
XVII, 3 p. 128) is nothing more than a mere guess, reminding of 
the theories of Ture Hederstrom. — F. Jonsson explains the word 
Singasteinn as »old stone» (Maal og Minne 1931 p. 148); he connects 
if with goth, sineigs. As no other traces of this word are left in Old- 
\orse (with the exception of the highly problematic sii]5steR on the 
runic stone of Tune) this explanation is likewise very uncertain. 

2 j)a, not j)ar. Again an argument against the topographical 
m 'aning of the preceding names. 

3 Flateyjarbok I, 275—283. 


130 Jan de Vries, The problem of Loki 1 FC 1 lo 

tion is added to it, telling about Sorli, who fought with 
Hogni Halfdanarson and who afterwards contracted a 
fostbroeSralag with him. Properly speaking this does not 
belong at all to the real subject of the story, although the 
I attr is named after this part. 

A second introduction, prefixed to the story, tells the 
following myth: Four dwarfs, called Alfrigg, Dvalinn, Ber- 
lingr and (frerr, make a precious jewel. Freyja wants to 
have it, but must pay for it with one night of love spent 
with each of the dwarfs. Loki, who knows about this scandal¬ 
ous bargain, tells it to Othin, who, however, as the only 
reward for his information compells him to steal the jewel 
from Freyja. In the shape of a fly he enters into the well- 
locked room of the goddess, whom he finds sleeping with 
the jewel round her neck. As a flea he gives her a sting, 
whereupon she makes a movement by which the necklace 
can be unfastened. Freyja, missing it the next morning, 
claims it back from Othin who wishes to return it only 
on this condition, that she succeeds in bringing two mighty 
kings into an everlasting battle. 

Freyja executes this order by means of a curious device. 
She changes herself into a woman, named (fendul, and sits 
down at three different times at a place, where she is sure 
to meet HeSinn. The first time she advises him to fight 
Hogni: the result is that they rather unexpectedly swear 
blood-brotherhood. The second time she gives him a po¬ 
tion of oblivion and suggests to him the abduction of Hugni’s 
daughter. Finally she gives him a narcotic to lay upon him 
the alog of the everlasting fight. 

This part of the I attr is rather confusing and badly 
composed. It is obviously pure invention to account for 
the spectral battle, in Christian times supposed to be the 
result of a curse, only to be taken off by the most Christian 
king Olaf Tryggvason. The author of this romantic story 
makes great use of heathen adornments; he knows i hat 

FFC llo Chapter VI, Loki and the Brxsingamen 131 

Othin in those former days fixed the supreme hour of death 
for every hero and that it was the valkyries who had to 
execute the orders of the Lord of Battle. But he does not 
know what has been once the real underlying idea of this 
belief; he makes Gondu] into a vile sorceress and he explains 
the everlasting battle by a curse, an alog, emanating from 
(ithin. 1 

But the people in this autumn of Old-Norse civilisation 
were not so easily satisfied. They were Unable to accept 
the explanation of these matters of ancient lore by a quickly 
understanding belief; so, like curious children they inquired 
about all the particular details of this queer curse. Story¬ 
tellers easily hit upon some explanation or other in the 
storehouse of mythological antiquities: the witch in the 
v'ood had some reason for lying this curse on the fighting 
heroes and Othin also had some reason for compelling her 
to do so. A combination with the ever quarrelling Othin 
and Freyja was ready-made; a myth was grafted upon 
the heroic tale, although in such a poor and careless fashion, 
that a real unity did not ensue. 

The myth of the theft of the jewel surely was not in¬ 
vented by the author of the pattr; in this case we might 
expect a better and more logical coherence between the 
different motives. On the other hand it is not probable 
that he left the original tradition untouched; it is more 
likely that he remodelled it according to thirteenth-century 
taste and that he sadly spoilt it. Hence the question: which 
part wars added afterwards and what was the primitive 
form? We only can make a guess, and this in a most loose 
and haphazard way. The trick of Loki has been long re¬ 
cognised as the motive of a w T ell-knovn folk-tale (Aarne- 
Thompson Nr. 560), although its peculiar form in the j)attr 

1 Cf. if. O’. Boer, Untersuchungen iiber die Hildesage in the Zeit- 
scheift fiir deutsche Phiologie XL, p. 12—19. 

Jan de Vries, The problem of Loki 


FFC 110 

shows strange deviations from the ordinary form. 1 If we 
subtract this as a later borrowing the bare fact remains 
that Loki once stole a precious object in a cunning manner. 

It is Freyja’s breastjewel, that Othin wished to possess 
and that Loki without any obvious reason is compelled to 
fetch for him. We observe that this myth does not tally 
at all with the ordinary scheme of such tales: here Loki has 
to repair the wrong he has previously done, whereas in the 
Jiattr he has only had the imprudence to indulge in scandal¬ 
ous gossip. Moreover he is not compelled to fetch an object 
for Othin, that has been lost, but a thing the stealing of 
which may be expected to gravely vex his compromised wife. 

This can not be the original structure of the myth. But 
our effort to penetrate into the core of this puzzling tale of 
scandal does not yield the secret of its proper meaning. 
On the contrary, what we retain are the poor and scattered 
fragments of an irretrievably lost mythical world. It is an 
example of conjugal dissidence between Othin and his wife, 
known from the days of the Langobards; it is Loki scheming 
some trick; it is finally a divine jewel made by some dwarfs. 
But whenever we wish to grasp these motives in their real 
mythological meaning, we are deluded: the ruse of Loki 
is an ordinary folkloristic motive and the dwarfs bear names 
that easily betray the fact of their being invented. Dvalinn 
has been borrowed from the Edda, Alfrigg from continental 
literature, Berlingr and Grerr most probably from the 
realm of some author’s fancy. We may safely finish this 
unpromising investigation by saying that for the study 
of Loki this tale is unable to yield any valuable informa¬ 
tion. This Loki is the stereotyped trickster of so many 
fabulous stories, only somewhat clumsily acting out of his 

1 Cf. Panzer, Hilde-Gudrun p. 164 and E. O■ Sveinsson, Verzeich- 
nis islandischer Marchenvarianten, FFCommunications \r. 83 p. 
XXXVII sqq. 

FFC 110 

Chapter VI, Loki and the Brisingamen 


2. T h •* myth of t It e B r l s i n g a m e n 

If we combine all these various details we arrive at this 
conclusion: Loki and Heimdall have been fighting for the 
possession of a precious piece of amber, endowed in loth 
century tradition with poetical terms as Singasteinn and 
I'dgasker. Snorri tells us that it properly belonged to Freyja 
and that its ordinary name is Brisingamen. On the other 
hand there is a very fantastic tale about a jewel of Freyja, 
but where its proper name is not mentioned. How did 
Snorri know that this brooch was named Brisingamen? 
He mentions two kennings where this word occurs; in ch. 16 
of Skaldskaparmal the expression pjofr Bnsingamens for 
Loki, in ch. 20 eigandi Brisingainens for Freyja. But we 
are not entitled to conclude that Snorri actually had found 
these kennings somewhere in a poem, that has been lost 
afterwards. They are of a rather prosaic character. Ken¬ 
nings with the present participle eigandi such as he freely 
mentions in the last named chapter ( eigandi Valfalls ok 
Sessrumnis ok fressa , Bnsingamens ) are nowhere to be 
found, neither in Eddie nor in skaldic poetry, unless in 
the thirteenth and fourteenth centuries 1 II . Hence Snorri 
has not found these figures in actual poetry; he has compo¬ 
sed them himself. We must bear in mind, that the poetical 
expressions he mentions in his manual for skalds are only 
partly drawn from aut hentical sources; a considerable amount 
has been invented by him to show how a special type might 
be used. 

We can not but admit that our material is rapidly mel¬ 
ting away before the fire of our inquiry. Again, where did 
Snorri get this name from? There is one, but only one poem 
where it occurs and, curiously enough, this is one of the 
latest Eddie poems, the I>rymskvi9a, where we read twice 

1 Cf. hugar eigandi, Skj. II B p. 268 and oglis tuns eigendr, Skj. 

II B p. 573. 


Jan de Vries, The problem of Loki 

FFC 110 

(st. 15 and 19) men BHsinga. As I have argued in the above 
quoted paper 1 , at the time Snorri wrote his treatise, this 
poem had not yet been incorporated in the collection of Eddie 
poetry; of course this does not imply that it did not exist; 
it may indeed have been composed some time before but 
not yet have arrived at such a venerable age that it was 
deemed worthy of figuring among the genuine tales of the 

Perhaps the reader may protest that as early as the 
tenth century we know from the Haustlong (st. 9) the ken¬ 
ning Brising girdipjofr for Loki. This does certainly prove 
that Loki as the thief of such a precious object was known 
centuries before Snorri. But a Brisings girdi is not the same 
as a Brisingamen. Neither is the singular Brisingr an 
equivalent of the plural Brisingar, nor can we say that a 
belt and a jewel are synonymous. 

From Old-Norse tradition we learn the following facts: 

1. Loki is the thief of Brisings girSi (Haustlong, 9th 

2) Heimdall and Loki have a quarrel about a piece of 
amber called Singasteinn (Husdrapa, 983) 

3) Loki is the thief of a precious collar of Freyja’s, that 
bears no particular name (Soria] attr) 

4) The jewel of Freyja is called Brisingamen (f>rvms- 
kvida, Snorra Edda). 

Of course we may piece this information together in 
this way, that the Brisings girdi is the same as the Brisinga¬ 
men. that furthermore the piece of amber, mentioned in 
the Husdrapa, properly belonged to Freyja, that lastly the 
author of the Soria Mttr has omitted the name of Freyja’s 
jewel, although it was known as Brisingamen. Moreover we 
may argue, that if Loki has thrice stolen a precious jewel, 
this object must have been the same in all three cases. 

1 Tijdschrift voor Nederl. Taal- en Letterkunde XLVII, p. 296. 

FFC no 

Chapter VI, Loki and the Brisingimen 


But we may take the tradition also in a different way. 
If Loki has thrice stolen a precious jewel, it must he granted 
that it is probable, that there exists a close connection be¬ 
tween these tales, but it does not follow, that they were 
practically the same. They may have been developments of 
an original motive, that in course of time took very diffe¬ 
rent forms. In fact, they have done so, for obviously a 
battle in the sea is not precisely the same as Loki’s trick 
in the shape of a flea. Here we observe the results of mere 
story-telling: Loki as the thief of a jewel was a theme that 
could be elaborated in different ways. 

This is why I prefer to take the four tales as the succeed¬ 
ing stages of the literary treatment of a motive. The heathen 
traditions (Nr. 1 and 2) tell about a belt called Brisings 
gir&i and a piece of amber called Singasteinn. There is not 
the faintest trace that they ever belonged to Freyja. In 
much later time we find a quite romantic tale about the col¬ 
lar of Freyja. ’s, which has been stolen by Loki; the form in 
which it is told does not tally with the usual conception of 
this god; we get the impression that it is pure fiction. This 
jewel is called in sources belonging to the 12th and 13th 
centuries Brisingamen. We do net know, how old this 
name is. 

I must begin with emphasizing the fact that the available 
tradition gives no indications regarding the mythical mean¬ 
ing of the jewel. That it should have indicated the sun is a 
mere supposition, for wich the affirmed fire-nature of Loki 
seems to be the strongest support. Much in his paper on »Der 
germanische Himmelsgott» proposes to consider the myth 
as an intruder upon an older story about the theft of fire. 
As an argument in its favor he mentions the disguise of 
Loki as an aquatic animal, here a seal and in an other tale, 
which we will presently discuss (§ VIII, 1), a salmon. But 
the conception of Loki as a fire-demon in this latter tale is 
also highly questionable; so we can not use it as an argu- 

136 Jan de Vries, The problem of Loki FFC HO 

ment for the same (supposed) character of the god in a 
quite different myth. 

Much has compared the name Brisingamen with on. 
brisingr, norw, brising »fire», the verb brisa »to flame, to 
glitter» m. icel. brisheitr »qui percaluit». But of course this 
does not prove either that this jewel has been substituted 
for a spark of fire, or that it might have meant the sun. It 
would be a no less intelligible name, if it simply indicated 
a »brilliant jewel» x . 

Hugo Pipping has in one of his excellent treatises on 
Eddie mythology 1 2 submitted the view, that the Brisingamen 
originally means »the aurora borealis». Accordingly he ex¬ 
plains the word as jewel of the Brisingar, which is to be 
accepted as a plural 3 . The Brisingar are the flames of the 
aurora borealis, in fact valiant men fallen in battle. This 
opinion seems to me quite unacceptable. If according to U. 
Holmberg 4 with the Tlingits of Alaska and the Laps the au¬ 
rora borealis is supposed to be a heavenly flock of warriors, 
fallen in battle, this does not imply that the idea of Valholl 
necessarily must go back to this natural phenomenon. There 

1 The suggestion of R. Eisler , Weltenmantel und Himmelszelt 
p. 161 that it perhaps might be connected with the collar of stars 
belonging to the Dea Mater, venerated in Minor Asia (cf. the jewel- 
studded collar of the Istar Belit-Ilani), which the learned author more¬ 
over himself calls an unsettled question, finds as far as I see, no sup¬ 
port in the Old-Norse tradition. 

2 Studier i Nordisk Filologi XVII, 3, pp. 126 sqq. 

3 In note 1 on p. 129 he says that the form Brisings in Haustlong 
has no value whatever against Brisingamen to be found in the 
])rymskvi6a 12 and in the Snorra Edda. But here it is not the oppo¬ 
sition of one example against two, but of a very old source against 
two recent traditions. So we can not discard the form Brisings so 
easily. And if one does not accept my opinion about the late date 
of the J>rymskvi9a, the proportion of one source with the singular 
name and one other with the plural form does not prove the greater 
reliability of either of them. 

4 Valhall och Varldstradet, Finsk Tidskrift 83, p. 337 sqq. 

FFC 110 

Chapter VI, Loki and the Brisingamen 


are indeed more examples of the belief that only the brave, 
who died in war, went to a heavenly abode or that they were 
gathered to a, distinct place. 

This belief is wide-spread among the Finno-Ugric peop¬ 
les. Holmberg quotes an old account of the Ostiaks religion \ 
where it is said: »If the beasts of the forest tear one asunder, 
or if one is shot in battle, his soul goes upward, but the souls 
of those dying a natural death at home go downward)). 
He quotes some other examples and mentions in this connect¬ 
ion the belief about the aurora borealis 1 2 . It is not quite 
certain if this belief is so very old; Karjalainen is of opinion 3 
that heaven as a dwelling-place of the dead is only to be 
found among the more southernly tribes and that it is of 
tartarean origin. Hence the connection with the aurora 
borealis may be also recent. At any rate without this motive 
we find a corresponding belief among many other peoples. 
Among the Fiji a soul can not be admitted to the Happy 
Land before he has proved that he has died a violent death. 
Otherwise he must go back to the upper air and die respec¬ 
tably i.e. violently 4 . The same holds good for the Mortloc-k- 
islanders: the souls of those who fell in battle are carried by 
the war-god Rasim to his special heaven, where the combat 
is continued 5 . The Aztecs believed also that the warriors 
fallen in battle went to a special heavenly abode, the house 
of the sun. The same is said about the Indians of Central 
America: those who die at home go into the earth, but those 
who die a violent death go upwards to Tamagastad and 
Qipattonal, where the sun rises 6 . Obviously Frobenius is 

1 The Mythology of all Races IV, p. 80—81. 

2 See also his examples about Siberian tribes, ibidem p. 488. 

3 Die Religion der Jugra-Volker, FFComm. Nr. 41, p. 189—190. 

4 J. E. Harrison, Prolegomena to the Study of Greek Religion 
p. 575. 

5 Sir James Frazer, The Belief in Immortality III, p. 119. 

6 IF Krickeberg, Marchen der Azteken und Inkaperuaner, Maya 
und Muisca pp. 31 and 213. 


Jan do Vries, Tiie problem of Loki FFC 110 

right when he says 1 , that in a warlike tribe death on the 
field of battle implies a happy life in the other world and it 
is not surprising that we find the same belief among the 
heathen Scandinavians and the Oaraibs, as well as in many 
other places of the globe. 

The explanation of Pipping is moreover based upon the 
supposition that the tale of the SoriaJiattr represents an 
original myth, especially in as far as the Hjadningavig 
should have been a direct consequence of the theft of the 
Brisingamen. If this jewel is lost, the aurora borealis has 
disappeared. To get it back a new slaughter of warriors 
must take place to act as flashes of light. This is indeed very 
ingenious, but one may ask why the warriors who died at an 
earlier time and hitherto formed the aurora borealis have 
disappeared all of a sudden. What is the relation between 
the jewel and the fallen warriors, both being mythical re¬ 
presentations of this same polar phenomenon? 

So it is not clear to me why the Brisingamen should be 
identified with the element of fire and how the story of 
the fight between Loki and Heimdall in the shape of seals 
can be reconciled with the conception of the jewel as the 
aurora borealis. I consider it to be a dangerous method to 
connect the superstitions of primitive peoples with very 
recent and romantic sagas in order to arrive at an original 
meaning of a mythical representation the traces of which 
are not to be found in the oldest sources. 

Therefore we may ask if this precious jewel indeed has 
any definite mythological meaning. Hans Jungner shares 
the same doubt 2 ; the goddess Freyja possessed of course a 
necklace just as any mortal woman and only in course of 
time it became incumbent upon mythographers to find a 
special explanation for this ornament. The original form, 

1 Erlebte Erdteile III, p. 62. 

2 Cf his book Gudinnan Frigg och Als harad, p. 141. 

FFC 110 

Chapter VI, Loki and the Brisingamen 


however, is the necklace, not the very complicated myth 
about it. This seems sound reasoning and therefore we 
might ask, if it could not he somehow connected with the 
Brosinga mene of the Beowulf, which Pipping is obliged to 
deny. 1 

Many years ago 2 , Sophus Bugge drew our attention to 
the singular fact that there is a curious coincidence between 
the Hama (= Heimir) of the Anglosaxon poem and the god 
Heimdall in the Husdrapa. This may be more than a simple 
whim of chance. It is indeed rather surprising that Heimdall 
figures as an active person in a mythical tale and besides 
that as an opponent of Loki. It is not to be denied that a 
purely heroic tale about Hama, who fled from Eormenric 
with a precious jewel (as Walther and Hildegunde escaped 
from this tyrant with large treasures) may have been cast 
to the mould of a mythological tale. Still the differences are 
great. If Hama’s jewel is a ring, in the Husdrapa it is a piece 
of amber. In the Beowulf nothing is said about a struggle 
for its possession, which on the contrary is the principal 
motive of the Husdrapa. We might observe that the brief 
mention of the jewel in the Old-English poem by no means 
exhausts the whole story about it and that a Scandinavian 
poet of the 10th century may have heard more and fuller 
information than is available to us. But it is clear that we 
are apt to become entangled in the thicket of endless con¬ 
jectures if we wish to proceed on this way. 

Finally how must we explain the queer alternation of 
the words Singasteinn and Brisings girdi (or Brisingamen )? 
The Anglosaxon word Brosinga mene, transmitted to Old- 
Horse tradition may have gradually changed into a form 
like Brisingamen, which had at any rate the advantage of 
being somewhat intelligible. Several scholars have refused 

1 L.c. p. 129. 

2 In Paul und Braune’s Beit rage XII (1887) p. 69 sqq. 


Jan de Vries. The problem of Loki FFC 110 

to acknowledge the identity of the words on account of the 
difference between the vowels; but then they only took into 
consideration the possibility of an original connection be¬ 
tween these names. If, however, the Old-Norse tale is an 
adaptation of an Anglosaxon heroic story, we can expect 
that arbitrary changes have taken place in the period of 
transmission; of this fact we have indeed several instances, 
the most conspicious being that of Sigfrid and SigurSr. 1 

Did Ulfr Uggason somewhere in his long account about 
the fight between Heimdall and Loki make use of the name 
Brisingamen? It is not impossible, but nothing more can 
be said. What we know is this: he names the jewel with 
skaldic kennings, such as Singasteinn and Vagasker. The 
first name is of course not to be explained as a mere shorten¬ 
ing of the real name. But some sort of pun may be intended: 
did he choose this kenning, because it unmistakebly reminded 
one of the real name Brisingamen? If so, we may suppose, 
that the word was rather arbitrarely composed, since it 
was not so very easy to find an expression at the same time 
different from and reminiscent of the real name. I venture 
to suggest that a word from an Anglosaxon source has slip¬ 
ped into the skaldic poem: sigle ond. sincfcet the Brosinga 
mene is called in 1. 1200 of the Beowulf and the word sine 
as well as its numerous compounds are of common use in 
Old-English poetry. Has a word sincsldn, which actually 
occurs in Anglosaxon, been imitated by a Norse poet and 
slightly adapted to a form more congruent with the proper 
name Brisingamen? 

I should not like to take this idea for more than a possi¬ 
bility. Still it may show us once more, how grievously 
little we know about many so-called myths of the Old- 
Norse tradition, indeed so very little that there seem to be 

1 Cf. A. Heusler, Zeitschrift. fur deutsches Altertum LII, p. 
97 sqq. 

FFC no 

Chapter YI, Loki and the Brisingamen 


no limits for hypotheses that may be constructed in every 
possible direction. 

In the myths of the Brisingamen Loki is a mere name 
Has he ever been more than a name? Has Loki ever had in 
any original form a real role, by which his character was 
clearly defined? We must confess that we are absolutely 
ignorant in this respect. I wish to sum up the results of 
this chapter in the following way: 

1. In the Soria} attr Loki is the cunning trickster, but in 
a debased and untraditional form. He is the hero of 
a simple folk-tale. 

2. The fight between Heimdall and Loki is utterly 
obscure as to its real meaning. It may be that it 
never had any mythical meaning at all especially if 
it is an adaptation of a foreign tale. For the study 
of Loki’s character it is a worthless source. 

3 The fight in the water and the disguise in seal-shape 
are likewise quite incomprehensible. It would be the 
utmost arbitrariness to conclude on the base of these 
unreliable mythical tales that Loki stood in some 
particular connection with the element of water. 



1. Loki in his relation to Othin and Thor 

In his paper »Myterne om Loke» Axel Olrik has tried to 
make a clear distinction between two different types of 
Loki. In a few traditions he is closely connected with Othin 
and Ha nir; we have treated them in our IYth chapter. As 
Olrik accepts the identification of Loki and Lodurr, he 
credits to this group: the myths of Andvari and I&unn, the 
creation of mankind, the Lokkatattur and the English magic 

In other mythological tales Loki is the companion of Thor. 
The examples are the tales about Utgar&aloki and Genre3r 
as well as the 1 rymskviSa. We might ask if this distinction 
implies a different character of the god? In answer to this 
question Olrik invests a list of myths according to the funct¬ 
ion of Loki. 

1. He tricks the trolls. Examples: Lokkatattur. the 
birth of Sleipnir and the myth of Andvari. Although 
he has nearly the same character in the J>rymskvi8a, 
it may be argued that this belongs essentially to the 
Loki of the Othin-group. 

2. Loki falls into the power of the giants and is obliged 
to free himself. Examples: the myths of pjazi, Geir- 
ro5r, Andvari. Although two of these traditions show 
him in the company of Othin and only one as the 

FFC llo Chapt. \ 11, The character of Loki in o.-n. mythology IW 

follower of Thor, Olrik ventures to conclude that this 
Loki should have been more particularly the servant 
of the thundergod. He considers the Esthonian myth 
to belong to the Thor-Loki group. 

3. Loki as a thief. Examples: the myths of the Brisinga- 
men and of Sifs hair; moreover the Esthonian myth. 
By making a slight addition Olrik shows that this 
character of Loki also belongs to the Thor-Loki group. 

After having divided the extant material in this way, 
olrik tries to combine it with a geographical distinction: 
Othin-Loki is found on the Faroes, but Thor-Loki »in its 
typical puritv» appears in the Esthonian myth; hence the 
conclusion is rather easy: Thor-Loki belongs to a more 
eastern, Othin-Loki to a western group. Olrik concludes his 
argumentation with the exclamation: Now the problem is 
solved, why Loki has a double position in the world of the 
gods: two religious traditions, differing both in character and 
in place, have been mixed up in Eddie mythology 1 . 

These rather rash conclusions show the weak points of 
Olrik’s method. Our investigation has led us to the rejec¬ 
tion of the Esthonian myth as a reliable tradition and at any 
rate as unrelated to the Scandinavian Loki-tales. About 
the originality of the story in the Lokka-tattur we enounced 
our serious doubts. Two pillars of Olrik’s system are broken 
down. It is curious to observe, that Olrik in his endeavour 
to explain the difficulties of the Eddie tradition has sought 
help from popular tradition, nay he has even give more credit 
to the Faroese ballad and the Esthonian folktale (in Kreutz- 
wald’s text!) than to the most venerable relics of Old-Norse 
literature. Indeed, again a strange overestimation of popu¬ 
lar traditions, which are considered to be so important that 
the Eddie tradition may be judged by it. 

I do not wish to follow this way. Olrik arrives only at 

1 I eslskrift Feilberg - , p. 551—554. 

Jan de Vries, The problem of Loki 

FFC no 


his conclusion by a rearrangement of the material. The 
types Othin-Loki and Thor-Loki are not clearly distinguished 
by the character of their myths; on the contrary in real tra¬ 
dition they seem to occur in the same kind of tale. This 
weakens the validity of Olrik's theory; the method, however, 
is not sound either. If it is true, that Loki presents himself in 
two quite distinct characters, then this must become evident 
from the myths, told about him, and it is to be expected 
that the Old-Norse sources give more valuable material 
than later distorted popular traditions. The myth of 
Sleipnir’s birth does not stand on the same level as the 
Lokkatattur nor the myth of f>jazi on that of the Esthonian 
folktale. Again, if we make a distinction into three types 
of Loki’s character, they must not overlap each other; but 
they do this in Olrik’s system. The Andvari-myth shows the 
types 1 and 2, moreover at the same time the Othin-Loki 
and the Thor-Loki: the Esthonian tale belongs likewise to 
the types 2 and 3. The I5unn-myth disappears altogether 
from the scheme, as does the tale of Utgar&aloki. In fact, 
there is here no distinction in the character of Loki, as he 
is found in the three groups of Olrik: Loki as a trickster of 
the giants and as a thief apparently is the same sly, mischie¬ 
vous deity. 

The reasoning of Olrik is quite arbitrary. The different 
myths, cleverly arranged, do not show a clear tripartition, 
but Olrik has forced the mythological material into the mould 
he has prepared beforehand. His starting point was the 
undeniable fact that Loki is allied to Othin as well as to 
Thor; but he has made the mistake of using all the traditi¬ 
ons from heathen times onward to modern popular lore 
(even of a foreign people) without making any clear distinct¬ 
ion for the reconstruction of the original types. 

FFCUO Chapt. VII, The character of Loki in o.-n. mythology 145 

2 . The double character of Loki 

If there are two or even more types we must make it clear 
in what way they differ from each other. This does not de¬ 
pend upon the motives of the mythological tales; Loki as a 
thief may be the same deity as the trickster of giants. And 
secondly, if there are two or more types it does not follow 
that they are distinct in time and place. Again I must 
insist upon the fact, too easily forgotten, that the historical 
analysis of the material is not the most important task of 
our investigation, but that we must try to arrive at an under¬ 
standing of the mythical traditions as religious phenomena. 

Is there indeed any difference in Loki’s character, as he 
appears in the myths, we have hitherto discussed? We may 
begin with a short survey of the motives, occurring in these 

In clipping Sif’s hair Loki makes mischief against a god¬ 
dess without any reason (Sif 1 ). The ensuing tale of the con¬ 
test between the dwarfsmiths presents him as compelled to 
repair his fault (Sif 2 ), but it is again of his own free will 
that he entices the dwarfs into a contest (Sif 3). In the 
myth of the giant-builder he gives advice that goes wrong 
(Finn l); then he is compelled by the gods to make it good 
again and he plays a trick on the giant (Finn 2 ). In the tale of 
I J j azi, he begins with an attack upon the giant-eagle, that 
takes the meat from him and his companions (jazi 1 ); 
then the giant compells him to steal IQunn ( : jazi 2 ) and 
finally the gods induce him to bring her back (. jazi 3). 
In the myth of GeirreSr he has the imprudence to venture 
into the neighbourhood of a giant (Geirro&r 1 ); consequently 
he must free himself by bringing Thor without his weapons 
to the giant’s home (Geirr 08 r 2 ). On this journey he is nothing 
but a mere companion of the god (GcirreSr 3). In the tale of 
Utgar 8 aloki he is likewise only the humble servant of the: 
thundergocl. The myth of the Brisingamen presents him 



Jan de Vries, The problem of Loki 


as being compelled by Othin to rob Freyja (Sorli). The prose- 
introduction of Reginsmal states, that he kills the otter 
without any reason (Andvari 1); later on he is sent away to 
catch Andvari (Andvari 2). In the i>rymskvi8a he is the 
cunning companion of Thor and succeeds in deceiving the 
giants. Finnally in the Lokkatattur he acts quite independ- 
antly of Othin and Hoenir. 

We may tabulate these different motives according to the 
following principle: 

A. 'The main distinction is, if he acts voluntarily or under 


1. Loki acts voluntarily 

a. to the detriment of the gods (Sif 1, Finn l) 

b. to the detriment of the demons (Sif 3, f>jazi 1, 
Andvari 1, Lokkatattur) 

c. without any definite purpose (GeirraQr 1) 

2. Loki acts under compulsion 

a, compelled by the gods Q)jazi 3, Andvari 2, Finn 
2, Sif 2, Sorli) 

b. compelled by the giants Q)jazi 2, Geirra5r 2) 

3. Loki acts as a mere companion (GeirraSr 3, t>ryms- 

kvi5a, UtgarQaloki). 

This method of classification shows at any rate, that 
Loki acts under some sort of compulsion in more cases than 
of his own free will. The preponderance of the former group 
is really much greater than is shown by our tabulation, for 
among the examples under 1 a and 1 b there are some doubt¬ 
ful cases, wich perhaps ought to be removed. In the begin¬ 
ning of the tale about the giant-builder Loki gives advice that 
turns out badly, although he obviously has not the intention 
of injuring the gods; he was not compelled to do so, but as 
the gods are deliberating about the decision to be taken, the 
frame of this tale corresponds with that of the examples 
under 2 a. With regard to the tales, grouped under lb, the 
Lokkatattur must be dropped as altogether unreliable. 

FFC no Chapt. VII, The character of Loki in o.-n. mythology 147 

Andvari 1 is clearly a case of reckless violence; he kills the 
otter, hut does he know that this will have such serious 
consequences? Again the case Jjjazi I shows slight differences 
from the type lb; nobody compels him to beat the eagle, but 
at any rate he does it on behalf of his companions. Moreover 
he only defends his right and is unaware of the danger he 
exposes himself to. Hence the only clear cases of Loki doing 
some mischief purposely are Sif 1 and Sif 3. We have already 
shown that this chapter of the Snorra Edda is very loosely 
constructed and we have no certainty whatever that it 
exactly renders the motivation of the original story. In 
fact Sif 1 is of no importance: when a myth tells that the hairs 
of a goddess are made by the dwarfs, then it is only natural 
to suppose that the real hair by some accident has been des¬ 
troyed; cf. the Irish myth about the silver hand of Nuadu. 
The malefactor is easily found in Loki, but the arbitrariness 
of the conclusion is shown most emphatically by the fact, 
that Loki has no reason whatever to act in this brutal way: 
Loki Laufeyjarson hafdi ]>at gort til Icevisi, at klippa har alt 
af Sif. 

In the second group Loki rather often acts under the 
compulsion of the gods; in most cases he has to atone for 
some mischief (jazi 3, Andvari 2, Finn 2, Sif 2). When he 
is constrained by the demons, he can only free himself 
from captivity by yielding to their wishes. In such a case 
he is not to be held fully responsable for the mischief done. 
This leads us to the question: what is his relation to the gods 
and the demons? The following classification may demon¬ 
strate it: 

B. Loki’s relation to the gods and the demons 

1. Loki acts to the detriment of the gods 

a. voluntarily (Sif 1, Finn 1) 

b. under compulsion (frjazi 2, Geirrodr 2, Sorli) 

2. Loki acts to the detriment of the demons 


Jan de Vries, The problem of Loki 

FFC 110 

a. voluntarily (J)jazi 1, Sif 3, Andvari 1, Geirro&r 1, 

b. under compulsion (Finn 2, Sif 2, Iqazi 3, Andvari 2) 

c. as a mere companion (GeirroSr 3, I>rymskvi8a, 

This result is surprising. Only 4 cases under Bl against 
12 under B2. Moreover the examples under 1 are not to be 
judged equally. If Loki does a mischief to the gods under 
some sort of compulsion, this is of course meant as an excuse 
for such monstrous treachery. On the other hand he will 
not needlessly put himself into the power of the demons; 
hence he must be compelled in this case also. The reason 
for the compulsion in the cases lb and 2b are absolutely 
different. Now, if the cases under lb show Loki only reluct¬ 
antly doing evil against the gods, the most important cases 
for his enmity against them are Sif 1 and Finn l. I may 
refer to my previous discussion: Sif 1 is an arbitrary addition 
of some myth-monger (or of Snorri?) and Finn 1 belongs 
probably rather to the same exceptional cases as Andvari 1, 
for he acts inconsiderately and is thereby the cause of serious 
danger to the gods. But it is he again, who has to avert it, 
and he does it to the entire satisfaction of the gods. 

We arrive at the following conclusions: 

1. Loki is more often a foe of the demons than of the 
gods. In the latter case he is represented as the victim 
of the consequences, for being captured by a troll he 
can only free himself from death by complying to its 
demands. On the other hand he shows his enmity to 
the giants, sometimes without any apparent reason; 
in those cases, where he is compelled to acts of violence 
his reluctance is explained by his fear for the danger 
he has to face. 

2. The motives of his deeds are not the same in all the 
examples. If he is constrained to these actions, this 
obviously does not prove anything either for or against 

FFC110 Cliapt. VII, The character of Loki in o.-n. mythology 149 

his character. In one single case he acts from pure 
malignity, hut Sif 1 is no real myth at all. More 
commonly he acts without considering the possible 
consequence of his deed (Andvari 1, pjazi 1, Finn 1, 
Geirrodr 1). As far as the structure of the story is 
concerned, this is decidedly an extremely poor motive 
and it does not do credit to the sagacity of Loki, howe¬ 
ver much this is extolled in other tales. Certainly it 
is an important fact that all these cases occur in the 
introductions of the myths, that is in those parts where 
additions and arbitrary changes are most common. 
From a mythological point of view it shows the indif¬ 
ference of Loki’s deeds as to his relation to the various 
classes of supernatural beings. He does not act with 
any definite purpose, he does not place himself in a 
position of enmity against either gods or giants; he is 
more of a mischiefmaker, who avails himself of any 
possible opportunity and who is himself afterwards 
much surprised at the extent of the injury. 

3. There is no reason to make a distinction between dif¬ 
ferent conceptions of Loki. He is always the same per¬ 
son, but the tricks he plays show him constantly in a 
different position towards the gods and the demons, 
the two opposite powers between which the fate of 
the universe is in suspension. Hence it may seem as if 
he were a conscious actor in this world-drama; indeed, 
he is the only being that deals blows to both sides and 
Loki has consequently the appearance of being a 
deserter from the camp of the gods. We will presently 
have the occasion to discuss this view, which, however, 
although clearly taken in a group of myths, is quite 
out of place here, 

On the whole the sources for these myths are rather 
recent. Most of them are only known from the Snorra Edda 
(Sif, Finn, GeirroQr, UtgarSaloki). The Andvari-tale is told 


Jan de Vries, The problem of Loki 

FFC 110 

in the prose-introduction of the Reginsmal; we do not 
know at what time this exclusively Scandinavian accretion 
to the Nibelung-legend has been composed. Only the myth 
of t>jazi is of venerable age: it is told in Haustlong and conse¬ 
quently it was fully developed in the 9th century. Here we 
find also an allusion to the theft of the Brisingamen, al¬ 
though we are unable to decide if it belongs to the type 
of the SorlaJjattr or of the Husdrapa. At any rate we are 
warned by this important and fully reliable example, that 
the character of Loki, as we have described it, was quite 
developed at the time, when the very beginning of literary 
tradition sheds its faint light on the Teutonic past. 

But our previous investigation has revealed to us the 
baffling fact, that the tale of pjazi, even in its oldest at¬ 
tainable form, is not mythical in the proper sense of the 
word. There are too many simply folkloristic motives, too 
much composition and arrangement of originally inde¬ 
pendant elements, whilst on the other hand the real religious 
»meaning» is obscured by the literary treatment. This is, 
of course, still more the case in those tales, that are only 
known from the Snorra Bdda. Here obviously centuries 
of pure myth-making, only inspired by literary predilec¬ 
tion for queer stories of ancient lore, have distorted the 
real features. But, happily, this does not matter very much. 
For the character of Loki is sharply fixed and it remains 
in its constitutive elements invariably the same. Moreover 
the oldest attainable form in the Haustlong tallies com¬ 
pletely with all the succeeding stages: Loki is not a deadly 
foe of the gods, on the contrary he is their companion and 
even in some respects more daring and active than Othin. 
But then the consequences are terrible for him and he is 
compelled by unforeseen circumstances to rob the gods 
of their most precious possessions. In fact he is the sport 
of Fate, that tosses him to and fro between the vengeance 
both of gods and giants. 



1. Discussion of the sources 

In our introductory chapter II we have paid attention 
to different theories about the proper meaning of Loki; 
the hypothesis of his being a fire-demon gained the greatest 
number of adherants. Still the evidence for this character 
of the god is extremely slight and the old texts are at any 
rate not quite explicit. The following arguments have been 
adduced in favour of this opinion: 

1. the etymology of the name Loki, 

2. the myth of his being caught in the shape of a salmon, 

3. the theory about his being the special companion of 


4. the popular tradition about a fire-demon called Lokke. 

As in questions of the interpretation of religious phe¬ 
nomena etymology is an unreliable guide, we might begin 
our discussion with the second point. In the 49th chapter 
of the G-ylfaginning we find the following story about 
Loki after the death of Baldr. Loki fled from the gods and 
hid himself in a rocky place, where he built a house with 
a door in every wall. Sometimes he took the shape of a 
salmon and sported in the falls of Franangr. Thinking over 
the devices the gods could possibly hit upon to catch him, 
he made a net, hut then all of a sudden the Asir approached. 
He flung the net into the fire and leapt as a salmon into the 
river. The gods were unable to find him. but the cleverest 

152 Jan de Vries, The problem of Loki FFC 110 

of them, Kvasir, saw the traces of the meshes of the net in 
the ashes on the hearth and so the gods contrived to make 
a net themselves. But Loki eluded them for a long time, 
til at last Thor succeeded in seizing him between his fingers; 
that is why the salmon is thin behind. 

Several scholars have asserted the close relation be¬ 
tween this curious tale and a Finnish magic song about the 
origin of fire. 1 This magic poem 2 consists of different parts, 
distinguished by Kaarle Krohn in the following way: 3 

A. The fire was rocked in a golden craddle, but fell down 
owing to the carelessness of its guardian, 

B. The spark fell into the lake Aloe or the Aloe-sea, 
which rose all in a foam and then dried up. 

C. The spark was swallowed by a fish; a net was hastily 
made by passing it through the water with the cur¬ 
rent and against the current the fish was caught; it 
was cut open and the spark was found in its inmost 

The third episode has some resemblance to the Old- 
Norse myth of Loki; this is explained by Krohn by the 
supposition that they both go back to a lost medieval 
catholic legend. The fact that no definite Christian tradi¬ 
tion can be pointed out as the actual source, weakens of 
course the validity of this hypothesis; at any rate we are 
obliged to rely exclusively upon the internal evidence of 
our sources. As I am of opinion, that the connection be¬ 
tween the Scandinavian myth and the Finnish magic song 
is all but proven, it will be necessary to enter into the details. 

1 Already Weinhold in Zeitschrift fur deutsches Altertum VII, 
p. 19. 

2 Printed in Lonnrot’s Suomen Kansan muinaisia Loitsurunoja 
(Helsinki 1880) p. 336—340 »Tulen synt,y», cf. K. Krohn , Suomalai- 
set syntyloitsut (Helsinki 1917) p. 100—131, translated as Magische 
Ursprungsrunen der Finnen, FFCommunications N:o 52 p. 109—145. 

3 Cf. F. Ohrt, The spark in the water, FFComm. N:o 65 p. 3. 

FFC 110 Chapter VIII, Loki and the element of fire 153 

We may for convenience sake begin with the myth of 

It consists of different episodes, to wit: 

1. The frame of the tale; the flight of Loki after the 

murder of Baldr, 

2. The invention of the net and the cleverness of Kvasir, 

3. The capture of Loki, 

4. The impression of Thor’s hand on the salmon. 

We may, of course, at once dismiss the last point as a 
simple aetiological legend well known in differents parts 
of Europe. Usually the grasp of a supernatural being serves 
to explain darker spots or lines on the fish. 1 

A much more difficult problem is presented by the 
second episode: the invention of the fishing net. Widely 
diverging views have been submitted to account for it, but 
on the whole scholars are inclined to consider it as a later 
addition to the story of Loki’s capture. A. G. van Hamel 2 
is of opinion that »it is not much like the cunning demon to 
provide his enemies himself with the means to seize him». 
Indeed, but on the other hand it may be thought a moving 
example of the irony of fate, as the most wily amongst the 
gods is betrayed by the faint traces of his own cunning. 
The gods were unable to hit upon such an ingenious idea; 
their enemy invented it for them and became the victim 
of his own invention. Of course, the motive itself, the first 
invention of the fishing net, does not originally belong to 
the myth of Loki, but its function in the tale as it appears 
in the Gylfaginning, is quite natural and does not show in 
itself that it could have superseded an other and older 

This is one of the reasons why I am unable to accept 
the construction of Van Hamel in the above quoted paper. 

1 O. Dahnhurdt , Natursagen I, p. 201 and II, p. 180—183. 

2 Neophilologus XIV, p. 207. 


Jan de Vries, The problem of Loki FFC 110 

He supposes that originally Loki should have been captured 
by means of the net of the seagoddess Ran. He bases this 
opinion upon ch. 31 (33) of the Skaldskaparmal, where 
Snorri explains the kenning eldr JEgis for »gold». The reason 
of it is the famous banquet in the hall of iEgir where all the 
gods are assembled as is described in the Lokasenna. Very 
acutely Van Hamel remarks that Snorri adds one phrase 
that properly speaking does not belong to the explanation 
of the kenning, for he says at the end: pa urdu cesir pess 
varir at Ran atti net pat er hon veiddi t menn alia pa er a, see 
komu. According to Van Hamel there is but one possible 
solution: the mention of the net presupposes that it played 
a r61e in the tale of Loki’s capture: the gods, assembled in 
iEgir’s hall, and planning how to catch the murderer of 
Baldr, see this excellent implement and borrow it from the 
sea-goddess. This is indeed a very plausible solution; it is, 
however, not the only possible one. 

Why should Snorri not have added a small detail to a 
story, he only quotes to explain a kenning, not for the sake 
of its contents? This chapter is a theoretical exposition of 
different sorts of poetical language; he is not at all interested 
in the myth itself. When he says: Ran er nefnd kona Mgis 
en niu deetr peira , sva sem fyrr er ritat , we are not allowed 
to conclude from it that these nine maidens were present 
at the banquet. So it is with the net. Snorri has such an 
amount of knowledge about heathen mythology, which 
he wishes to note down in his handbook, that he may easily 
unconsciously add a detail, not because it belongs to the 
story he is referring to, but because it furnishes a new piece 
of evidence for his treatise on poetics. 

On the other hand Celander is of opinion 1 , that the 
invention of the net is a very original feature in Loki’s 
history; the genuine character of it seems to be supported 

1 L.c. p. is. 

FFC 110 Chapter VIII, Loki and the element of fire 155 

by Swedish words for the spider’s web, where we find the 
name of this god: lockanat, lockasnara. This is a question 
of method, to which I will return, when I wish to discuss 
the folkloristic traditions about Loki. But the combina¬ 
tion of a net and the spider’s web is so self-evident, that 
it may appear more than once. In a description of the 
Ojibway superstitions the Jesuit missionary Allouez says 1 
about their god Michabous, »que ce fut dans ces Isles (Michili- 
makinak), qu’il inventa les rets pour prendre le poisson, 
apres avoir considere attentivement 1’araignee dans le 
temps qu’elle travaillait a sa toile pour y prendre des 

In my opinion two motives belong indissolubly to each 
other: the net and the fish-shape of Loki. If the net, in 
some way or other, is the most original motive, than Loki 
must appear in the shape of a fish; if on the other hand the 
myth related that Loki changed himself into a salmon, 
then of course it is most natural to catch it by means of a 
net. Now Loki as the inventor of a net is a motive abso¬ 
lutely unparalleled in Northern mythology: he never in¬ 
vents an implement useful to mankind. But he very com¬ 
monly assumes different shapes, and it does not matter 
which form he takes, if it is only fit for the aim he pursues. 
We have mentioned his disguise as a horse, a seal, a flea 
or a falcon. He seems to be at home in every element. 
Hence it is quite unwarrantable to conclude from the tale 
about his capture that he has some particular connection 
with the element of water. Moreover the myth is not 
necessarily of high antiquity, because the elaboration of his 
punishment seems to belong to a later stage of development. 
Then a mythographer may have combined different motives 
in order to compose an interesting tale. 

If this is the case, the most important feature of the 

1 The Jesuit Relations 54, p. '200. 


Jan de Vries, The problem of Loki 


myth is the shape of the salmon. It was to be expected 
that Loki should change himself into an animal, that is 
not easily captured. 1 Perhaps the red colour of the salmon, 
especially of the char (salmo salvelinus) may have led to 
the idea of some connection with fire or lightning. Dahn- 
hardt mentions an example of the belief, that salmon are 
afraid of thunder; then they lie motionless in the water. 2 
Thor being the professional pursuer of the demons, it may 
be surmised, that Loki changed himself into a salmon, 
because popular belief knew, that there was a contrast 
between this fish and the thundergod. On the other hand, 
if Loki had already been brought into some connection 
with fire, as we found this clearly expressed in the myth 
of Thor’s visit to UtgarQaloki, then the red colour of the 
salmon might easily suggest the idea of Loki especially 
taking the shape of this fish. In the short rendering of the 
myth by Snorri the red colour is not particularly mentioned, 
but if this fish belonged to the land-locked kind 3 , its colour 
was known to every one. A fish of a red colour in connec¬ 
tion with fire is found in the Finnish magic song »Tulen 
synty»; we meet with a curious parallel among the American 
Indians, where fire is extracted from a red salmon. 4 But 
I do not want to take this as a mere suggestion; I am fully 

1 Popular tradition in Norway still knows the rhyme: 

Dersom eg var saa god te’ hojre 
som eg er te’ sjaa 
saa skulle ingen mann paa Lande 
meg faa. 

Cf. Joh. Th. Storaker, Naturrigene i den norske folketro, Norsk 
Folkeminnelag XVIII, p. 267. 

2 Natursagen II, p. XII. 

3 I refer to the kind, known in England as char, in Germany 
as sabling or rotforelle, in France as omble chevalier. The Norwegian 
name is kolmunn (salmo carbonarius, not to be confused with the 
kolmule or gadus melanostomus). 

4 U. Holmberg, Aika 1918 p. 31. 

FFC 110 Chapter VIII, Loki and the element of fire 157 

aware of the fact, that by the word lax in the Gfylfaginning 
nothing is said about the particular species and that as 
Prof. Liestol had the kindness to tell me, nowhere in Nor¬ 
way has the popular belief about a connection between 
salmon and fire or thunder been found. 

It is, however, the supposition that the Loki-salmon 
is a kind of fire-fish, that the alleged connection with the 
Finnish magic song is based. Krohn has treated this ques¬ 
tion in a very careful investigation and he gives as its result 
the following possible similarities 1 : the linen net, the ashes, 
the reiterated fishing of the salmon and the fear of the sea. 
Those similarities are, indeed, very frail: if it had been 
established beforehand that Loki is a fire-demon and con¬ 
sequently his salmon-shape a manifestation of his nature, 
then we might lend some colour from this fact to these 
meagre correspondences. But since the conception of Loki 
as an embodiment of fire would find one of its strongest 
supports in a close relation between this myth and the 
Finnish song about the origin of fire, we must very care¬ 
fully weigh the importance of these possible similarities. 

The difficulty of catching the fish is a most natural 
motive and it seems to me of particular importance that 
in the details there is no connection whatever between the 
Scandinavian and the Finnish poems. The three efforts 
of Thor and the way in which Loki escapes the net, drawn 
through the river, are quite unknown in the »Tulen synty», 
where we find as the cause of the failure of the first efforts, 
that the net has been drawn awkwardly. 2 It is likewise 
of very slight importance that the material of the net is 
the same: lingarn in the Norse myth and in a Finnish variant 
of Ingermanland: ydlla Hina liylvettihin. The magic qualities 
of linen may account for the use of this material. 

1 He says very cautiously: mahdollisina yhtalaisyyksina. 

2 Cf. the words: nurin nuotta potkittiin, vaarin veettihin apaja. 

158 Jan de Vries, The problem of Loki FFC 1 10 

Furthermore Krohn mentions »the ashes». If we have 
in mind the interesting motive in the Scandinavian tale, 
where Kvasir sees the traces of the burnt net, we can not 
but feel that this argument, if it really were solid, might 
be sufficient to prove the connection between the two 
traditions. Unfortunately this is not the case. The Finnish 
song does not know this very peculiar motive and when it 
tells about ashes, it is in quite a different sense: the ashes 
are the result of a woodland being burnt down in order to 
get a soil fit for cultivation: a well-known primitive way 
of tilling the ground. In a variant from Karelia the ashes 
are not mentioned at all 1 , but sometimes we find lines 
such as: 

hara vanha poltettihin 
siihen liina kylvettihin, 

where the ashes, at any rate, are presupposed and in one 
variant of uncertain provenience, where a certain being, 
called Tarsilainen, is burnt in his iron boat (a not original 
motive in this magic song!) the poem continues: 

sen kypenet kylvettihin 
peltohon perittomahan 
maahan manterettomahan, 
siita kasvoi kaunis hamppu, 
pensi pellavas hikesi. 2 

There is, indeed, no connection between the motive of the 
ashes in both traditions. Whereas it has in the Gylfaginning 
the very singular meaning of furnishing the pattern of the 
net, it is in the »Tulen Synty» only an unimportant acces¬ 
sory of the burning of the forest. 

1 Krohn , Suomalaiset syntyloitsut p. 113. 

2 Ibid. p. 114. The words mean: The glowing ashes were sown 
on a field without borders, on a land without bottom; from it 
grew beautiful hemp, spread flax in all directions. 

FFC 110 Chapter VIII, Loki and the element of fire 


Finally the fear of the sea. In the Snorra Edda it is said, 
that it is dangerous for salmon to go into the sea: var Pat 
lifshaski at hlaupa a see,inn. Krohn combines this with the 
Aloenmeri, mentioned in the Finnish formula, which he 
explains as an equivalent of the Dead Sea. In the Christian 
formula, that might have been the source of both the Scandi¬ 
navian myth and the »Tulen Synty». the scene of the action 
was the river Jordan and the Dead Sea, But is it probable, 
that the idea of the curious nature of this sea could have 
been preserved during the various migrations of this Chris¬ 
tian magic song throughout Europe and finally should 
have survived even in its adaptation to a heathen myth? 
In fact, we must try to find more natural explanations for 
this curious detail. If Loki had transformed himself into 
a fresh-water salmon, we might connect the fear of the sea 
with the fact that this salmon never goes outside the rivers. 
But this is after all not less hypothetical than the opinion 
of Krohn. We have, however, other and simpler ways of 
explaining it; the motive is a necessary complement of a tale, 
where a fish is to be caught in a river: one might ask why 
Loki did not try to escape into the infinite ocean and this 
was indeed a very decisive answer. One might even suppose 
that it is the last trace of a more detailed form of the myth, 
where the mouth of the river had been obstructed by the 
gods in order that Loki could not escape that way. 1 If there 

1 The Scandinavian tale is not very clear. First the gods drag 
the net throug the river, Loki escapes by hiding himself between 
the pebbles on the bottom. Then they make a second effort upp 
til forsins (hence landwards), but Loki sees at skamt var til ssevar 
and leaps over the net upp i forsinn. We get the impression that 
the gods went not landwards but in the opposite direction. The 
second lime Thor himself wades through the river ok jar a sva ut 
til stevar. But then the gods must have known that here escape 
for Loki was impossible. Or did the tsde not take so much trouble 
about these details, prompted by our intellectual way of reasoning 
and is it only a later redactor who tried to smoothe away the incon¬ 


Jan de Vries, The problem of Loki 

FFC 118 

are so many possible solutions for this unimportant detail 
we need not recur to the Aloenmeri of the Finnish magic 

When we compare the myth of Loki and the »Tulen 
Synty» there is but one motive common to both: the making 
of a net. This is, however, in either case a rather sub¬ 
ordinate element of the tale. For the comparison of differ¬ 
ent folktales we must constantly bear in mind that the 
similarity of one single feature is not sufficient to prove 
the identity of these tales; this is a principle of all sound 
philological criticism. But if, as in the present case, the 
motive is moreover found in two widely different forms, 
there is no reason whatever to suppose that they should 
both have been derived from a non-existant original tradi¬ 
tion viz a medieval Christian legend. 

The case would be different if we might consider both 
poems to express in one sense or other a mythical concep¬ 
tion of the element of fire. If Loki as a fire-demon took a 
salmon’s shape, we might suppose, that this salmon had 
also a peculiar connection with fire. The idea of fire, how¬ 
ever, being concealed in the salmon is not only nowhere 
expressed, but there is no place for it in the structure of 
the story. It is possible that Loki has been, perhaps on 
account of the similarity between his name and the word 
logi, conceived of as a kind of fire-demon and that this 
was the reason for choosing the red-coloured river-salmon 
as a convenient shape for him, but this does not yet prove 
that the real significance of the myth has anything to do 
with the origin of fire. It is nothing more but another in¬ 
stance of myth-making in later times, when the heathen 
gods were a suitable subject for literary treatment. We 
must be cautious on our painful way of investigation lest 
we form too rash conclusions from such scanty material. 
On the one side we are threatened by the Scylla of seductive 
but very unreliable folklore traditions, on the other side 

FFC 110 Chapter VIII, Loki and the element of fire 161 

we are faced by the Charibdis of untraditiona], arbitrary 
myth-making by antiquarians, whose Christian faith is 
equally strong as their love for the glorious past of their 

2. The fire-nature of Loki 

Now it is time to resume our discussion of the arguments 
in favour of the hypothesis, that Loki originally was a 
fire-demon. We have cancelled the witness of etymology, 
we have tried to demonstrate that the myth of Loki’s cap¬ 
ture can not be brought into a close connection with the 
Finnish magic formula of the Tulen synty. The third point 
to be discussed is his relation to the thundergod Thor, 
which, however, I have not been able to demonstrate as 
an original and reliable belief (see ch. V, § 5). As to the evi¬ 
dence of modern popular belief, I will have the opportu¬ 
nity to treat this question at full length in chapter XII. 

Here we may draw our conclusions with regard to the 
Old-Norse material. And these conclusions can only be 
negative. The myths, where the fire-nature of Loki is 
accepted by the majority of scholars, are but a very frail 
base for such a hypothesis. Neither the contents of these 
myths nor the particular form of their motives are suf¬ 
ficiently explained by the fire-theory. In my opinion the 
real meaning of these obscure tales remains unintelligible, 
if we take Loki to be a fire-demon or not. If this is the case 
for instance in the myth of Loki’s capture we may ask: is 
the salmon-shaped Loki the thief of a spark of the heavenly 
fire? If the gods succeed in catching him, does it mean that 
he is only a very unsuccessful Prometheus, as the fire after 
all comes back into the possession of the gods? And more¬ 
over, what is the reason of the fire-god being persecuted by 
the gods; has it anything to do with fire’s relation either 
to mankind or to the gods? Or else is Loki only a malicious 
demon, whose original connection with fire is of no im- 



Jan de Vries, The problem of Loki 

FFC 110 

portance for the right understanding of this particular 
myth? The fact, that we are able to put these questions 
without the possibility of getting any satisfactory answer 
from the theory that advocates the fire-nature of Loki, 
tends to prove that the evidence is but exceedingly slight. 

On the other hand I do not wish to deny that we may 
get ever and anon a faint indication of some tendency to 
such a conception of Loki. I have accepted it without any 
hesitation in the case of the contest between Loki and Logi. 
There may possibly be more examples of it. But they do 
not prove that his character from the very outset is that 
of a fire-demon, nor even that he has been clearly under¬ 
stood as such in the heathen period. If there is no single 
myth, that is based exclusively upon this fire-nature or 
can be only explained by it, we must arrive at the con¬ 
clusion that the few cases, where he shows any connection 
with fire are only accidental accretions to his mythical 

One may ask, what is then the reason that Loki although 
originally not a fire-god, afterwards became considered as 
such? In the myth of Thor’s journey to Utgardaloki it is 
clearly the great resemblance between the words Loki and 
Logi that has brought about the eating contest of the god 
and the giant. Here, however, the fire-nature of Loki is 
not clearly defined. On the contrary he loses because he 
is inferior to the voracious element, that consumes the meat 
and the vessel at the same time. An other explanation 
might be afforded by the fact that he belongs partly to the 
demons, even to those who are most dangerous to the 
existence of the world and the supremacy of the gods. Hence 
it will now be appropriate to consider those tales where he 
is pictured more like a demon than a god. 



1. Loki in the myth of Balder’s death 

According to the account in the Snorra Edda it is Loki 
who by directing the missile of the blind Ho 5r causes the 
death of Baldr; on the contrary there is no trace of his 
activity to be found in the tale Saxo Grammaticus gives 
in his Gesta Danorum. Two possibilities present themselves 
to account for this curious fact; on the one hand we may 
assume that Loki is an intruder upon the myth, only added 
to it in the West-Scandinavian tradition; on the other hand 
he may have played a role from the very beginning and 
then he has been dropped in the more romantic form of 
Saxo, who made a pure heroic tale from an original myth. 

The advocate of the former opinion is E. Mogk; the 
latter view is supported by both G. Neckel and F. R. Schro¬ 
der. Theoretically they are equally possible and besides 
the testimony of our sources we must rely upon the ad¬ 
missibility of the conclusions arrived at on the base of this 
hypothesis. Neckel has in his monography on Baldr tried 
to explain this myth by means of a far-reaching influence 
of religious cults and representations current in Asia Minor. 
Whether we have to explain the similarities between these 
traditions as mere parallels, or as the consequence of his¬ 
torical influence, is a question that Neckel has not clearly 
settled; I am of opinion that the distance between Iceland 
and Asia Minor is rather great, even if we accept the Gothic 


Jan de Vries, The problem of Loki 

FFC 110 

people as a connecting link. Hence it is not very probable, 
that a rather complicated religious myth should have 
travelled so far. Moreover the explanation, Neckel gives 
of the transmission, is highly problematic: the poet who first 
created the legend of Baldr was a man who heard two 
myths among the Goths of the Donau, not from a Gothic 
but from a non-Germanic (perhaps a Thracian) narrator, 
viz one myth about Atys and another about Ishtar; this 
man has composed two different poems which he recited 
at a South-Scandinavian court. 1 

We may put forward the following objections. 1. It would 
be the result of pure chance that these Asiatic myths spread 
to Scandinavia. If this hypothetical poet had only heard one 
single myth or if he happened to have been a less successful 
poet, the supposed chain of tradition would have been 
broken and no Baldr myth would have come into existance. 
2. The distinction, bet ween the migration of the cult itself 
and of the poetical traditions attached to it, is a serious 
drawback to this theory. Why should a skald himself not 
have arrived at a poetical treatment of a religious belief, 
that had won access long before the time he wrote, and 
why should it be necessary for a Scandinavian poet to travel 
again to the Black Sea in order to collect fresh material 
for the literary treatment of the myth? 3. If the state of 
affairs had been so complicated as Neckel suggests, it would 
have been impossible to deduce this series of events from 
the extant sources; this seems to me an overestimation of 
the possibilities of a philological comparative method. 

The way by which the cult came to the North is left 
by Neckel much more in the dark than that of the sup¬ 
posed poems. In fact, here I see a new and very serious 
objection to this theory. We are credited with accepting the 
fact that during a couple of centuries as the result of the 

1 Die Uberlieferungen vom Gotte Balder, p. 224. 

FFC llo Chapter IX, Loki as the enemy of the gods 


traffic on the so-called »Gothic road» right through Russia, the 
cult-forms of Attis had travelled to the North and that on 
this base a Baldr-cult had originated in Scandinavia. But 
a tale, once heard, may easily be retold elsewhere; that is 
what happens every day. How could such a tale, however, 
have been so impressive that the religious system'of the 
Scandinavians was affected by it? It would seem to demand 
a real cult, but a cult is quite different from a'mere tale. I 
do not believe that the Scandinavians of those days, having 
seen some thrilling religious performances in Eastern Europe 
and perhaps having been initiated into their mysteries, 
could have succeeded in introducing a new cult into their 
native land. This is particularly questionable, if, as Neckel 
himselfs asserts, this new cult showed a totally different 
character from that of the Scandinavian practices. More¬ 
over the curious resemblance to fertility-rites, still known 
in modern Teutonic folk-lore, suggests rather the idea of 
high antiquity and originality. If Eastern influence imposes 
itself by the weight of accumulated evidence, then I would 
be prepared to accept the fact that on an original religious 
representation expressed as well in cult as in myth, new 
elements of foreign extraction were grafted. But then it 
becomes again higly questionable if the meagre material 
of Old-Norse tradition will allow us to make a clear dis¬ 
tinction between the native and the foreign elements. 

But even if we accept Neckel’s explanation, what about 
the role of Loki? He is of opinion that the figure of the 
malicious enemy is ancient, but he is not able to find an 
exact prototype for him. On the whole the Egyptian Seth 
seems to come nearest to the Scandinavian traitor. At any 
rate he must come from Asia Minor, because the strong 
distinction between good and bad is typical of Persian 
belief. This argument may be true, but then it does not 
follow, that this non-Teutonic character of Loki has been 
developed under the Asiatic influence in the Baldr-myth, 

166 _ Jan de Vries, The problem of Loki _ FFC 110 

nor that the way of this influence should have been the 
»Gothic road». The Christian belief might have been of 
some importance too in this connection! 

Obviously the distance between the Icelandic Loki and 
the Egyptian Seth is too great for Neckel to connect them 
directly with each other. The German scholar F. R. Schro¬ 
der has tried to bridge this gap by supposing an intermediate 
form in Asia Minor. The resemblance, however, between 
Loki and Seth is only very vague, being mainly the mutual 
character of dangerous enemies of the gods, Seth killing 
Osiris and Loki causing the death of Baldr. As a further 
confirmation of his hypothesis Schroder reminds us 1 of 
the punishment of Seth according to the tradition of the 
Thebaic royal tombs that shows a greater resemblance to 
the myth of Loki; the likeness is not so very persuasive as 
it mainly lies in the fact, that they are both bound, a very 
usual punishment for a criminal. The office of Isis to watch 
the chain by which Seth is fettered, is quite different from 
the task of Sigyn, for Isis must prevent Seth from setting 
himself free, whilst Sigyn, filled with compassion for her 
disgraced husband, collects the ever falling drops of poison 
in a vessel. 

According to Schroder Loki was originally a god of 
vegetation (cf. ch. II), but he appeared at the same time 
as a fire-demon. Then this deity reminding one of the 
Greek Prometheus, was connected with the fettered giant 
of the Caucasian mountains (according to Olrik’s theory) 
and finally the influence of the Egyptian Seth (presumably 
of his Asiatic prototype) constitutes the last step in this 
evolution. By trying to conciliate the different theories 
about the origin of Loki Schroder has not succeeded in 
giving a coherent explanation of this enigmatical deity. 

Before accepting, however, this rather complicated 

1 Germanentum und Hellenismus p. 112 sq. 


FFC llo Chapter IX, Loki as the enemy of the gods 

development, let us study its real foundations. We can 
hardly do it in a more efficacious manner than by sub¬ 
mitting ourselves to the guidance of E. Mogk , who has most 
energetically denied, that Loki originally has had any role 
in the drama of Baldr. He has first pointed out in his lucid 
treatise about Snorri’s treatment of mythological tradi¬ 
tions 1 . that Loki is nowhere called the murderer of Baldr 
with the exception of the Snorra Edda; he has, criticising 
Schroder, afterwards treated the subject in a new paper 2 , 
where he puts forth the following arguments: 

1. In all ancient sources (such as Yoluspa, Baldrs draumar, 
HyndluljoS and Saxo Grammaticus) Ho Sr is exclusively 
the murderer of Baldr. 

2. There is no kenning in the total extent of Old-Norse 
poetry with an allusion to Loki’s role as the slayer of 

3. Only Snorri knows Ho Sr as the blind god. 

4. The prose-end of the Lokasenna does not combine the 
punishment of Loki with the death of Baldr. 

Hence the most important questions to be answered 
are: a. What is the form of the myth in the Voluspa? 
b. How are we to explain the Lokasenna? 

To begin with the Yoluspa. E. Mogk says that st. 
32—34 form an independant section, where the death of 
Baldr is related. It terminates with the formula Vitud er 
enn eda hval? Then a new scene is presented: the binding 
of Loki (st. 35) again finishing with the same sentence. I 
am very much convinced by Mogk, that this points rather 
to a succession of two quite in dependant myths. In the 
short survey of Baldr’s death we are informed that Ho Sr 
was the real murderer of Baldr; this is expressed most 
clearly by the words Hodr nam skjota and by the kenning 

1 Novellistische Darstellung mythologischer Stoffe Snorris und 
seiner Schule, FFComm. No 51, p. 11 sqq. 

2 I.okis Anteil an Baldrs Tode, FFComm. No 57. 

168 Jan de Vries, The problem of Loki FFC 110 

Baldrs andskoti. It is the same conception as in st. 9 of 
Baldrs draumar: 

Ho dr berr ha van hrodrbarm f>inig; 

hann man Baldri at bana verda 

ok Odins son aldri raena. 

Moreover the scene of the Voluspa is perfectly worked out 
to a conclusion; after this treacherous murder a terrible 
revenge is taken: the sun of Othin being only one night 
old kills Ho dr. In fact here is no place at all for Loki, he 
is not mentioned either in the perpetration of the crime or 
in the revenge for this odious murder. To the mind of a 
heathen Scandinavian the act counterbelanced by the 
revenge was in itself complete. And if Loki had been the 
real culprit, a poet like the author of the Voluspa might 
be expected to lay the whole burden upon him, not upon 
the obscure Ho dr. Prom this treatment of the tale we do 
not get the impression of a heinous villany, but of a lament¬ 
able catastrophe. The two opponents are Baldr, the valiant 
god and Ho dr, the incarnation of the battle. It might be 
supposed that f.e. a dioscuric pair of deities had fought 
against each other with the sad consequence that one slew 
the other, without violating the testimony of this source 
in the least. 

The mention of Loki in st. 35 must then belong to a new 
part of the poem. In fact it does. At st. 34 the volva term¬ 
inates the history of the past; from this moment onwards 
she will only speak about the preparation of the Ragnarok. 
She begins with an enumeration of the demoniac beings, 
who menace the cosmic order: first comes a description of 
the different places, where the trolls and the giants await 
the moment of their terrible attack. Foremost the horrible 
abode of Loki, then the dwellings of the giants and the feed¬ 
ing of new monsters by the old hag in the Ironwood. After- 

FFC 110 Chapter IX, Loki as the enemy of the gods 


this the signals for the final onslaught are given: the cocks 
in the world of gods and demons begin to crow. 

Indeed, the conclusion of Mogk seems well supported 
by facts and it is largely because we are accustomed to look 
through the spectacles of Snorri, that we are induced to 
make a connection between st. 35 and the preceding ones. 
It may be quite the contrary: Snorri came to the conclu¬ 
sion that Loki was punished on account of the death of 
Baldr, because he wrongly interpreted the series of stanzas 
in the Voluspa. His craving for logical order and causal 
connection made him seek for an explanation of the 35th 
stanza, where Loki is described as a prisoner. If this stanza 
describes a definite period of the god’s life, then we should 
ask: what is the reason of this severe punishment and the 
idea might easily present itself, that we could be able to 
find the answer in the preceding tale of Baldr’s death. But 
st. 35 is only meant to show the deadly foe of the gods in 
a place, where he awaits the day of his final revenge: it is 
the place and the personality not the history of Loki, that 
the poet wishes to mention here. If Mogk is right in his 
assumption, we must conclude that Loki has developed 
into an enemy of the gods quite outside the Baldr-myth 
proper and we are faced by the problem as to how this has 
come about. 

I do not expect that in such an exceedingly obscure 
matter as the problem of Baldr and Loki the combined 
arguments of Mogk and the present author will convince 
those scholars, who hold an other view, but at any rate I 
will assume that the conclusive force of this part of the 
Voluspa in favour of the opinion, that Loki, even according 
to this poem, has been the principal actor in the drama of 
Baldr, is somewhat weakened and can not be used without 
any restriction. 

What requires a closer investigation at present is the 
Lokasenna. Mogk contends that in the prose at the end 


Jan de Vries, The problem of Loki 

FFG 110 

of this poem the fettering of Loki is represented as a punish¬ 
ment for his sneering at the gods. This is of course question¬ 
able as this prose tale need not belong originally to the 
poem itself. The testimony of the st. 27—28 is of more 
importance. Here Frigg says in answer to a very insulting 
remark of Loki: 

Veitstu ef ek inni settak iEgis hollom i 
Baldri likan bur: 

tit Jm ne kvsemir fra asa sonom 

ok vseri \)k at t>6r reibom vegit! 

Thereupon Loki retorts with the following stanza: 

Enn vill M, Frigg at ek fleiri telia 

mina meinstafi: 

ek I)vi red er t>ti rida serat 

sidan Baldr at solom! 

It is accepted commonly that this poem supposes Loki 
to have caused the death of Baldr, but Mogk takes a differ¬ 
ent view of the question: Loki here makes an allusion to 
his preventing the return of Baldr from Hel by refusing to 
weep in the shape of the giantess Jokk. 1 Indeed, this is* 
quite possible, but it can not be proved. One might be in¬ 
clined to urge that the words of the stanza er J)u rida serat 
sidan Baldr at solom only mean his not coming back from 
the underworld. If Loki only wished to say, that he was 
the cause of Baldr’s death, he might have expressed him¬ 
self in the same way as Frigg did: »I am the cause that he 
is not present in the hall of iEgir». But he says: »It is my 
fault that you do not see him riding towards the hall» 
and so he lays particular stress upon the fact, that Baldr 

1 This has already been observed by Niedner, Zeitschrift fur 
deutsches Altertum XLI (1897) p. 306. 

FFC 110 Chapter IX, Loki as the enemy of the gods 171 

is not in the same position as the other gods, hut may he 
expected to arrive from an other place. Although I am 
inclined myself to take this explanation as the most natural 
rendering of this much discussed stanza, I am fully aware 
of the fact that it is only a possible interpretation, that does 
not even impose itself. Therefore I will cross over to the 
opposite standpoint and accept the fact, that this stanza 
is an allusion to Loki having been the radbani of Baldr. 1 

In a previous chapter (VIII) I have discussed the opinion 
of Van Hamel who has written the best and most exhaustive 
study about the prose frame of the Lokasenna . 2 Here he 
comes to the following conclusion: 

1. The gods meet in order to start the persecution of Loki 
who has caused the death of Baldr. Loki lives as a 
salmon in Franangrsfors, where he is caught with a net 
Loki’s punishment (Prose-frame, Gylfaginning ch. 49). 
2 a. The net used for the catching of Loki is the net of Ran 
the sea-giantess (Skaldsk. ch. 33; also in the original 

2 b. Later variant: The net was made by the gods them¬ 
selves, although they only imitated Loki’s own in¬ 
vention (Gylfaginning ch. 49). 

3. In connection with 2 a: the meeting of the gods is iden¬ 
tified. with one of their feasts in the hall of iEgir and 
Ran (prose-frame, Skaldsk. ch. 33). 

4 a. Loki ventures into the hall himself, where he slays 
Pimafeng, one of ACgir’s servants (Prose-frame, Skaldsk. 
ch. 33). 

4 b. Later variant: the slaying of Flmafeng is supplanted 
by Loki’s quarrel with the gos (Lokasenna). 

1 This does not imply that I accept the view of Neckel l!c. 
p. G1 and Schroder he. p. 110, that the expression ek pel red means, 
that Loki is not the actual perpetrator, but only the auctor intel- 
lectualis, for 1 agree with Mogk, FFC 57, p. 4 that the verb rada 
simply means »auctorem esse, efficere». 

2 Neophilologus XIV, p. 214. 

172 Jan de Vries, The problem of Loki FFC 110 

I can only admire this sagacious interpretation of the 
different traditions about Loki’s appearance in the hall of 
JEgir. It is moreover a logical development from a very 
simple beginning to a highly artistical ending; one might per¬ 
haps ask how it came to be, that at stage 4 a such a very 
unimportant detail as the slaying of Fimafeng was inserted 
and how this motive was connected with the punishment 
of Loki. But this is only a question of detail. The main 
result of Van Hamel’s study is, that we have to separate 
the poem itself from the prose-frame and that they both 
represent different traditions about Loki. 

Then we may ask, whether the Lokasenna necessarily 
supposes that the god is considered by the author as the 
murderer of Baldr and that he is punished for this crime. 
One might argue in the following way: the poem itself only 
tells about Loki’s appearance at a feast of the gods, where 
he insults the guests present and finally escapes after setting 
the hall on fire. It is not said that the gods are deliberating 
about his crime, nor that he is caught and fettered after¬ 
wards. Van Hamel has observed this with great acumen 1 ; 
the text of the poem implies that Loki is not seen in the 
hall until his unexpected appearance in st. 6. Nevertheless 
there must be some reason for Loki not being present at 
the feast: in st. 2 Loki is told by the cook, that the gods are 
boasting of their arms and prowess and that their words 
are not friendly to Loki. Moreover the crowning crime of 
Loki’s career, says Van Hamel, the treason against Baldr, 
has been perpetrated (according to st. 28) before the quarrel 
with the gods, which is the subject of the Lokasenna. 

The first argument, however, is not of much importance. 
At a time when Loki had gradually developed into the foe 
of the gods, whatever the reason may have been, a poet 
might have invented a scene, describing the gods assembled 

1 L.c. p. 205. 

FFC lln Chapter IX. Loki as the enemy of the gods 173 

at a meal and being of an unfriendly disposition towards 
Loki, who has not been invited. But it does not necessarily 
follow that the reason is the death of Baldr. Quite on the 
contrary, when Loki expresses his resolution to enter the 
hall, Eldir does not warn him because the gods are planning 
a terrible revenge, but because Loki by his sneering is likely 
to provoke them and induce them to an act of violence. If 
we read the second stanza in its context, we can scarcily 
interprete it in the way Van Hamel does, without adding 
that this is only a possible explanation and that it does not 
follow from the contents of the poem itself. 

Loki as the murderer of Baldr is inconceivable to me 
in this poem. How is it possible that Eldir does not leap 
into the hall announcing to the gods that the traitor has 
ventured into the lion’s den? How is it possible that the 
gods, filled with wrath against the murderer, whose punish¬ 
ment they are planning, calmly suffer him to enter and 
that no one dares to take hold of the criminal? We know 
the poor pretext of the poet: Loki’s appeal to the blood- 
brothership between him and Othin (st. 9). But then Othin 
could never have undertaken any act of violence against 
him! Why was there no other of the assembled deities to 
step in in Othin’s place? It was not dangerous to attack 
him, for he was ekki mikil voxtum according to the Sorla- 
liattr. Why had they to wait for the arrival of Thor? The 
prose-frame gives an other explanation: far var gridastadr 
mikill. Very inconsiderate of the gods, who are planning 
revenge for an atrocious crime, the more so because they 
had been prevented once from punishing Loki by quite 
the same impediment. But the idea is very interesting, as 
we find here a well-known situation: the gods are assembled 
in a gridastadr , when there comes an enemy to insult them: 
it is only Thor arriving from distant regions, who is not 
bound by the sworn oaths and who may use violence against 
the intruder. This is in fact the case of Hrungnir (Skald- 

174 Jan de Vries, The problem of Loki FFC 110 

skaparmal ch. 17), where we find exactly the same situa¬ 
tion; the gods are likewise prevented from taking revenge 
for his ofryrdi because they had given grid to him. The 
parallelism of the two stories makes it fairly probable, that 
the situation of the Lokasenna is analogous: the insulting 
of the gods is the essential matter and the peculiarly spicy 
detail lies in the fact that the gods have prevented them¬ 
selves from turning out the disrespectful boaster. 

But what about the stanza 28, where Loki confesses 
that he has caused Baldr to be absent? In my opinion, so 
far from being an argument in favour of Van Hamel’s 
theory, the stanza contradicts it clearly. We might expect 
that in a poem, treating of the vengeance of Loki, the acme 
of the story would be the moment when Loki himself boasts 
of his crime. Then the gods at last should have shaken off 
their strange torpor; they ought to seize their weapons, to 
shout at the impudent fellow and if they were not able to 
kill him, at any rate to take hold of him and to stop his 
insulting flow of words. Nothing of this kind happens. The 
tale runs quietly on. 

If we considere the structure of the poem, we may divide 
it into the following parts: 

st. 1—10 Introduction. Loki is given a seat among 
the gods, 

st. 11—56 The senna, 

st. 57—65 Loki is ousted by Thor. 

In the very middle of the senna Loki says to Frigg, that 
he has prevented Baldr from coming to the feast. What 
is now the effect of his words? No threatening words of the 
gods, no turning away in disgust from this rogue, not even 
the slightest idea of a feeling that now the cup of insult 
is filled to its very brim. Freyja meekly observes: 

orlog Frigg hygg ek at oil viti 
f)ott hon sialfgi segi. 

FFC 110 Chapter IX, Loki as the enemy of the gods 175 

If the author of the Lokasenna was a real poet — and no 
one will probably deny that — he could not have written 
this silly episode if he had had one moment in his mind 
the situation Van Hamel postulates for the poem. He had 
quite a different plan: a highly amusing scene in the family 
of the gods (not unlikely similar ones on venerable Olympus) 
where Loki showers his impudent remarks upon the divine 
beings, man had long ago learned to despise. The only 
thing, that Loki may be expected to do, is clearly expressed 
in st. 4: 

hropi ok rogi ef ]ju eyss a holl regin 
a ber muno bau berra bat. 

But Loki knows better: they are not even able to do this! 

Moreover when Thor comes, he threatens to smash Loki 
to pieces; and again we ask: how is it possible that he is 
allowed to escape? Why do they not make use of this 
opportunity to seize the criminal as Thor may do without 
breaking any oath? The poet has not felt this necessity: 
he only shows us in a deliciously comical scene, how Loki 
is again insulting the newly arrived and dangerous opponent, 
whilst poor Thor can not do anything but stammer again 
and again: 

begi bu, rog vsettr! \m skal minn brubhamarr 
Miollnir, mal fyrnema. 

Finally the poet himself clearly expresses the plan of the 
senna: it is the prolonged insult of all the gods by a sharp- 
witted Mephistopheles. Loki takes his leave, because the 
hammer of the thundergod obliges him to do so; but after 
all, he has had his say and even more than that: 

(st. 64) kvab ek fyr asom kvab ek fyr asa sonom 
bats mik hvatti hugr! 

Perhaps one might feel inclined to raise the following objec- 

176 Jan de Vries, The problem of Loki FFC 110 

tion to my explanation. St. 28 of the poem implies that 
Loki has perpetrated a mischief against Baldr; hence the 
senna takes place in the time between his crime and his 
punishment. This makes it fairly sure, that in one way 
or another the situation is about the same as Van Hamel 
presupposes: the gods are assembled to plan his revenge 
and in this moment he appears. 

But this conclusion is not obligatory. If we place our¬ 
selves on the standpoint of the poem (without the prose- 
frame), we only see the flyting scene of Loki: It supposes 
all kind of wicked deeds committed by the different gods. 
But it does not consider them in any chronological order; 
it does not show us Loki on a particular point of his carreer, 
when he has committed a series of crimes and revenge is 
awaiting him. If we accept such a chronological order, we 
are again under the impression of Snorri’s systematisation. 
Then we suppose, that for a poet the different myths about 
the gods are linked together by their mutual relation. This 
is not necessary; it is even very improbable. If a mythogra- 
pher tells us about the revenge of Brokkr, can he then fix the 
time when the lips of Loki were sewn together? Evidently 
not before the senna, for then there would have been a 
silent flyting. Between the senna and his final capture, 
however, there is no place for the episode of the dwarf. Or 
must we infer that the thread Yartari has been loosened 
and that Loki afterwards recovered his speech again? 

The myths are not co-ordinated, they do not form one 
continuous history of the gods. They belong to different 
spheres of religious and mythological tradition; they are 
very often in contradiction to each other. A poet may take 
from the tradition whatever motive he wants, without being 
compelled to consider at the same time the causes and the 
consequences of it. Our author wanted to find the fiercest 
insults, Loki could hit upon; when he spoke to Frigg, of 
course the death of Baldr was present in his mind and it 

FFC 110 Chapter IX, Loki as the enemy of the gods 


was particularly effective, because it was Loki himself who 
had played a part in it. But in the whole train of his thought 
this is only a motive, even a very subordinate motive, and 
the poet does not realise the exact position of the senna, he 
has invented, with regard to the premises of the Baldr-myth. 
This, indeed, lies beyond his scope. He has only found a 
very effective taunt, and afterwards he passes on to a new 
one. immediately losing the former one out of his sight. 

It seems to me that from this point of view the meaning 
of st. 28 is not of very great importance. If we accept the 
interpretation of Mogk, the stanza only supposes that Loki 
played the role of the giantess f)okk; then the sequence of 
events has been: 1. the death of Baldr by Ho5r, who acts 
on his own behalf. 2. Loki opposes himself to the returning 
of Baldr by refusing to weep. 3. The gods take revenge by 
killing the murderer and fettering Loki. This is, indeed, 
a possible reconstruction; Loki acts in his usual way by 
means of a cunning trick, by which he frustrates the fulfil¬ 
ment of the gods’ wishes. 

Afterwards the connection between stanzas 34 and 35 of 
the Voluspa and possibly also the combining activity of 
successive mythographers have enlarged the importance of 
Loki in the Baldr-myth; his preponderance in the Ragnarok 
may have had some influence too. So he became the real 
author of the crime and as Ho5r could not be ousted from 
his place, Loki was put at his side as the malicious counsel¬ 
lor. This is however not less a hypothetical construction 
than any other attempt to solve the riddle. 

If on the other hand we accept the theory, that st. 28 of 
Lokasenna presupposes the whole situation as it is described 
by Snorri, then our conclusion need not be changed in its 
essential parts. The development of Loki has taken place 
some time earlier than the literary activity of Snorri. But 
as I have pointed out in a previous chapter, we should not 
speak only of Snorri and his school, but also of his prede- 



Jan de Vries, The problem of Loki FFC 110 

cessors. We are ignorant about the exact time the Lokasenna 
was composed; at any rate it has been done by a poet, who 
lived in a time when the belief in the gods had altogether 
disappeared; in this Lucianic satire we even observe a 
supreme disdain for the most important heathen deities. 
Hence the eleventh century might be the terminus a quo and 
this period seems even rather likely as the Lokasenna gives 
the impression of having been inspired by hatred against 
pagan belief. But I will not stress this argument, for after 
all the senna is meant as a capital joke and if the reputation 
of the gods is torn to tatters, the fun is none the less great 
for it. Hence a poem like this may have been composed at 
any moment between the tenth and the thirteenth century. 
But if we take into consideration, that the skalds shun the 
use of kennings, containing the name of a god, in the period 
between 1050 and 1150 and that Loki himself 1 is not mentio¬ 
ned at all between 1000 and 1175, it seems more likely to me, 
that it belongs to the second part of the twelfth century. 

After all this does not matter very much. In any case the 
poem is written after the collapse of heathendom. The 
conception of Loki, we find here, is not based on the original 
belief, but it is the fruit of later speculations, much more of 
a literary than of a purely mythological character. His 
participation in the drama of Baldr is not necessarily an 
element of his heathen personality; it is the consequence of 
later development and arbitrary combinations. Still one 
thing becomes clear by it: the Loki of this myth is quite a 
different deity from the cunning trickster, we have met with 
in the mythological tales, discussed in our previous chapters. 
Now he is the wicked demon, intent upon the destruction 
of the gods. 

What may have been the reason of this development in 

1 His relatives are mentioned sometimes, but the cases of Fenrir 
and Hel are quite different from that of Loki. 

FFC 110 Chapter IX, Loki as the enemy of the gods 


malam partem? There are two possibilities, as far as I can 
see: 1. it is the result of a tendency lying in the character of 
the god itself: the trickster, who often makes mischief and 
is a serious nuisance to the other deities, is disposed to become 
a real enemy. 2. it is the consequence of foreign influences, 
by which this deity assumed the rdle of a malevolent demon, 
formerly unknown to the heathen pantheon. This question 
must, however, be left unsolved till we have discussed other 
examples of Loki’s devilish nature. 

2. Loki and the Ragnarok. 

According to the Snorra Edda Loki has caused the death 
of Baldr; he has been punished for it by being fettered in a 
subterranean cave and as the sworn foe of the gods he will 
be released at the time of Ragnarok. This is a series of 
events, so perfectly coherent and logical, that we are readily 
inclined to think that so the mythological conception must 
have been from the very beginning. 

But after having seen that Snorri is a clever arranger and 
combiner of the traditions, that he had collected, we suspect 
that the same may have been the case here. In fact, if we 
read the Eddie poem, known as Baldrs draumar or Yegtams- 
kvida, we do not find such an exactly fitting myth. On the 
question of Odinn who may be the cause of Baldrs death, 
the volva answers: 

st. 9. Hodrberr ha van hrodrbarm frinig: 

hann man Baldri at bana verda 

ok Odins son aldri rama. 

And when the god again asks, how Baldr will be revenged, 
she speaks about the birth of Vali, who will kill Baldrs 
andskota. No single word is said about the participation of 

180 Jan de Vries, The problem of Loki FFG 110 

Lokistill the poem mentions him as one of the most im¬ 
portant actors in the drama of the Ragnarok, for in the 
last stanza the volva says: 

Sva komit manna meirr aptr a vit, 

er lauss Loki liSr or bondom 

ok ragnarok riufendr koma! 

Hence the conception of the fettered Loki and his participa¬ 
tion in the general upheaval of the world may he quite 
independant from his connection with the myth of Baldr. 
What is the relation between his punishment and the Ragna¬ 
rok? One might suppose, that he was connected in some way 
or other with the final catastrophe and that afterwards he 
was supposed to have been fettered. But if this has not been 
originally the punishment for Baldr’s death — as we conclude 
from the YegtamskviSa — then there seems to be no plaus¬ 
ible reason for it. According to Axel Olrik 1 2 the combination 
of Loki and the Ragnarok is the consequence of his punish¬ 
ment, and his being fettered goes ultimately back to the 
Caucasian legend of the fettered giant. 3 In itself this can 
not be said to be altogether impossible; still it is a far cry 
from Iceland to the Caucasian mountains. Perhaps Christian 
influence may be taken into consideration also. The Gos- 
forth-cross, now commonly dated from the 9th century, 
shows on its West side the figure of a sitting man, whose 
hands and feet are bound together. Before him a female 
being holds an object, that probably may be taken as a 
cup or a vessel. 4 This scene has been formerly considered 

1 This is a new and very strong argument in favour of Mogk’s 
interpretation of Voluspa st. 32—35. 

2 Festskrift Feilberg p. 560. 

3 Ragnarokferestillingernes Udspring, Danske Studier 1913. 

4 Good illustrations in Stephens’ paper in the Memoires de la 
Societe royale des antiquaires du Nord 1882—4 p. 133 sqq; K. Krohn, 
Skandinavisk Mytologi p. 152. 

FFC 110 Chapter IX, Loki as the enemy of the gods 181 

as a representation of Loki’s punishment; the woman may 
be compared with the faithful Sigyn who is said to hold a 
vessel in order to collect the drops of venom that are con¬ 
tinuously falling on the fettered Loki. But since the cross 
has several other scenes, all belonging to biblical tradition, it 
is to be expected that the bound Loki and his devoted wife 
are not represented on this stone, but the persons of a 
Christian legend. Kaarle Krohn has ventured the suggestion 1 , 
that they are Herodes and Herodias; the vessel, held up by 
the woman, being intended to contain the head of John the 
Baptist. This explanation is not very convincing, but that 
the scene is more likely to be biblical than heathen, seems 
to be proved by an interesting Norwegian tradition preserved 
in Setesdal. 2 The old wooden church of Austad is said to 
have had doorposts, that were beautifully carved. In the 
lower part there was to be seen a kind of cartouche. As 
Skar relates it: »Der var Mannlik utskorne. Paa den eina 
Tavla var ein bakbunden Mann som laag; eit Kvende heldt ei 
skaal uppyver han; ein Mann stod til Skrevs yver han til aa 
rista han upp med Sverdet; men ein annan meinka han det». 
This has obviously nothing to do with the myth of Loki, 
but it is, either a Christian motive, or a scene taken from 
some popular tradition about an epical or perhaps a local 
hero. 3 

Hence it does not seem necessary to agree with A. Olrik 
in supposing that the wicked Loki is a new type in the Scan¬ 
dinavian pantheon, imported from the East; the way in 
which this is explained, is also open to serious objections. 
For Olrik argues that Loki had developed into a wicked 
demon among the Goths in the time of the migrations under 

1 Ibidem p. 153. 

2 J. Skar, Gamalt or Ssetesdal II, p. 32. 

3 The other cartouche is of the same character: four serpents 
attacking a man, also bound and lying backwards. The serpents are 
seen biting him in his breast, shoulder, stomach and legs. 


Jan de Vries, The problem of Loki 

FFG 110 

the influence of Christian ideas about the devil; afterwards 
the Caucasian legend would have added the new motive 
of the punished giant. This, however, constitutes a series 
of hypotheses which are of a very speculative nature. Is 
there any reason to suppose that the G-othic peoples have 
known the figure of Loki? 1 Have we any certainty about 
the time, when his development into a wicked demon took 
place? Is it indeed necessary to seek for an explanation 
outside Scandinavia? 

It must be admitted that the theory of Olrik has the 
advantage of affording a coherent explanation of this side 
of the Loki-problem. But even though this system makes 
the impression of being more-probable, still it will be advis¬ 
able to look for a solution that does not imply so many 
hypothetical suppositions. In fact, the idea of an opposition 
between the principles of good and bad is not clearly deve¬ 
loped in the Old-Norse religion; but there is from the very 
beginning the clear antagonism between the gods and the 
giants. It is not an ethical distinction, but the deities are 
the preservers and defenders of the cosmic order, whilst the 
demons are the hostile inhabitants of »Utgard» who try to do 
damage to the world. This division of the supernatural 
beings into friends and foes is in complete harmony with 
the conception of an aristocratic warlike society. The deve¬ 
lopment, however, into an ethical conception of this dualism 
supposes a different view upon the real value of life. The 
giants were not necessarily bad nor the gods righteous; in 
course of time this distinction became felt as the natural con¬ 
sequence of this antagonism. Does it follow that the reason 
was the influence of foreign religious conceptions? By no 
means. We might ask, how has it been possible that foreign 
ideas, based on a totally different conception of life, have 

1 The argument of the Esthonian myth about the thundergod has 
to be given up of course. 


FFC 110 Chapter IX, Loki as the enemy of the gods 

been introduced into the North? Only as literary motives? 
As new and interesting tales about divine beings? Or as 
part and parcel of a religious system? 

I am fully aware of the strong influence, which Christi¬ 
anity, especially by Irish intermediary, has exerted upon 
the Scandinavians. This is fairly certain, more certain at 
any rate than the connection with the religious traditions 
of South-Russia, the Caucasian mountains or Asia Minor. 
But the question must be put, if there have not been germs 
of this development among the Scandinavian peoples them¬ 
selves. In the complex of the myths about Baldr and Loki, 
I see as the principal element the consciousness of man of 
the existence of sin and consequently his craving for redemp¬ 
tion. This need not be the direct result of either Christian 
or Oriental influences. It may have arisen by the slow disinte¬ 
gration of heathen mentality, the breaking down of an out¬ 
worn society. Those men, who only put their confidence 
a matt sin ok megin were really on the verge of becoming the 
most fervent adherents of a religion that promised peace to a 
tortured soul. 

Rummer has said that this belief in one’s own power is 
not to be considered as reckless self-confidence 1 ; not as the 
religion of a self-possessed atheist, but as the result of a 
devout feeling of security (»frommes Heimatbewusstsein»). 
It is difficult to believe it. If Kummer quotes as a proof for 
his conception the story of Hakon the good 2 , we can not 
be easily convinced by it. For the words »the king does like 
all those who believe in their own strength and worship 
Thor» are meant as an excuse for his making the sign of the 
cross over the sacrificial horn. This is clearly a very awkward 
prevarication in this difficult situation and we must not 
conclude from it, that such was the ideal of the heathen 

1 Midgards Untergang (Leipzig 1927) p. 75. 

2 Heimskringla I p. 92. 

184 Jan de Vries, The problem of Loki FFC^ 110 

warrior. More commonly we read about men, wbo do not 
want to make sacrifices to the gods and only put their 
confidence in their own strength. 1 

Examples of this kind must not be pushed aside as mi¬ 
sinterpretations in later Christian times. On the other hand 
the English scholar Rcuth is nearer the truth, when he 
says 2 : »if this mood was as general as appears from the 
evidence there must have intervened in the religious 
history of Europe an epoch of fleeting but magnificent 
heroism, in which man sought to overcome terror and 
despondency by ignoring them». But we must add: it 
was growing despondency they tried to overcome, not 
the final victory over it. The old religious system had 
gradually lost its grasp on the people; the turbulant life of 
the Viking Age had torn them away from the narrow home¬ 
stead; the wild sea had tossed them from Norway to France 
or Russia and thence back again to Scandinavia; the scenes 
of murder and destruction had, after all, not passed through 
their mind without leaving any feeling of troubled recollec¬ 
tion and remorse. The contact with the Christian world, the 
surprising heroism of nuns and monks, facing brutal death 
with a cheerful hymn on their lips, had taught them to 
appreciate other virtues than those they were acquainted 

This might be sufficient to awaken the trouble in the 
soul of a man, who had lost the unshaken certainty of his 

1 Landnamabdk 1, 4, 4: heir fedgar vildu eigi bldta ok trudu k 
matt sinn. 

2 God, Man and Epic poetry II, p. 67. Van Hamel (Acta Philo- 
logica VII p. 265) considers those men as the conservatives who 
do not yet believe in the assistance of the gods ; but in the megin, 
they are able to actualise in themselves by means of magic. I con¬ 
sider this to be most unlikely, as the belief in personal gods reaches 
back into so remote periods, that the memory of a pericd before this 
development can hardly have lingered on til the beginning of the 
Viking period. 

FFC 110 Chapter IX, Loki as the enemy of the gods 185 

heathen belief. To throw away the draff of an outworn reli¬ 
gion, to take refuge in the undaunted recklessness of a man 
independant of gods and priests, is easily done. But is it 
heroism? Or is it weakness, that shrinks from confessing 
its craving for divine protection? It is impossible to make a 
definite statement as to the real disposition of the people 
in this period of transition. At any rate, sooner or later, the 
frail appearance of strength must break down and then the 
feeling of helplessness would necessarily arouse the longing 
for a reconciliation with the divine Lord of Life. Then man 
wants to come into personal contact with the gods; his soul 
becomes sensible of the problem of sin and righteousness. 
This religious revolution may have taken place long before 
the official introduction of Christianity; it may have thrown 
its shadow upon the last centuries of dying heathenism. 

It does not matter very much if Baldr comes from 
Thracia or the fettered Loki from the Caucasian peoples. 
The soil must be prepared where such a seed is sown. If we 
are aware of the necessity to seek the explanation for the 
growth of the religious conceptions like those of Baldr and 
Loki, of the awful consequences of sin and the necessity 
of redemption, of a final catastrophe and the coming of a 
new world, not exclusively in foreign influences, we may 
deem it possible that out of their own religious feeling 
the Scandinavians were able to arrive at these ideas. 

The problem of Loki requires a solution from this point 
of view. If we accept the possibility of a borrowing from 
Christian conceptions and Eastern legends, we must first 
try to answer the question, how it happened that out of 
numerous possible influences these passed into the Scan¬ 
dinavian religious system. The question may be put also in 
this way: is it possible to explain the development of Loki 
into a wicked demon on the basis of his original character? 
And secondly: are there traces for such a conception in the 
Old-Norse traditions, independant of the Baldr-myths 

186 Jan de Tries, The problem of Loki FFC 110 

and the legend about the Ragnarok, that point in the same 

3. The family of Loki. 

When Snorri in his Skaldskaparmal (ch. 16) asks, how 
Loki may be called in poetry, the answer is: sva, at kalla harm 
son Farbauta ok Laufeyjar , Nalar, brodir Byleists ok Hel- 
blinda, fodur Vdnargands — fat er Fenrisulfr — ok Jor- 
mungandz — pat er Midgardsormr — ok Heljar ok Nam ok 


This is a very rich development of family-relations, 
where his connections with demoniac beings is most con¬ 
spicuous. Before trying to settle, what the meaning of these 
connections may have been, we must, first make a distinc¬ 
tion between the elements belonging to different ages. We 
might expect them to be the result of the later development 
into a kind of devil, traces of which we have occasionely 
found in the preceding chapters. But in the oldest sources 
they have already made their appearance. 

To begin with the Haustlong, composed by ]>jo5olfr or 
Hvini in the 9th century; here we read the following ken- 
nings for Loki: Farbauta mggr (st. 5) and TJlfs fadir (st. 8). 
In the Ynglingatal of the same poet there appear as circum¬ 
locutions of Hel: Loka mcer , jodis Ulfs ok Narfa (st.7) and 
Byleists brodur mcer (st. 31). Finally in the Husdrapa of Ulfr 
Uggason (983) Loki is called Farbauta mggr (st. 2) and again 
logseims fadir in the [orsdrapa of Eilifr G-oQrunarson (about 
990 in st. 1). 

Hence in skaldic poetry Loki is designated as 1. the son 
of Farbauti 2. the brother of Byleistr, 3. the father of Hel, 
Fenrisulfr, Jormungandr and Narfi. It seems obvious that 
he is connected with beings, commonly considered as demo¬ 
niac, for if we are as yet unable to say, who Farbauti and 
Byleistr have been, on the other hand the character of the 

Ffc 110 Chapter IX, Loki as the enemy of the gods 1«7 

Fenrisulfr, the world-serpent and the goddess of the under¬ 
world is quite clear. 

The Eddie poetry adds some new kennings, besides 
furnishing fresh evidence for those already known. So Loki 
is called Byleists brodir in the Voluspa (st. 51) and Hyndlu- 
ljod (st. 40). In the Lokasenna we read ulfs fgdur (st. 10) 
and a new kenning Laufeyjar sonr (st. 52), which reappears 
in the jirymskvida (st. 18). A real catalogue presents the 
40th stanza of Hyndluljo5: 

01 ulf Loki 
en Sleipni gat 
eitt I otti skass 
fiat var broQur fra 

vi5 Angrbodo 
vi5 Svadilfara; 
allra feiknast, 
Byleists komit. 

Finally the Snorra Edda gives, as we have seen, a very 
complete list of those kennings. We must, however, not sup¬ 
pose that the examples, the learned Icelandic mythographer 
adduces, are all borrowed from existing poems; he only 
gives the models for these circumlocutions, and if we do 
not find brodir Helblinda or fadir Vanargands in any known 
poem, it does not follow that such a poem has ever been 
known to Snorri 1 . 

1 C.f. E. Mogk, FFComm. N:o 51, p. 11. 


f n: 1 10 

Jan de Vries, The problem of Loki 

We may tabulate tbe different family-relations of Loki 
in the following way: 



son af Farbauti 

Husdrapa , 

son of Laufey 


son of Nal 


brother of Byleistr 


brother of Helblindi 


husband of Sigyn 


husband of AngrboQa 


father of Hel 



father of Fenrisulfr 



father of MiQgarQsormr 


father of Sleipnir 


father of Narfi or Nari 


father of Ali 


| Eddie 
I poetry 






Lokasenna | 


HymiskviQa J 





Gylf 32 
Skin 16 
| Gylf 32 
Skm 16 
Gylf 32 
Skill 16 
Gylf 32 
Skm 16 
Gylf 32 
Skm 16 
Gylf 32 
Skm 16 
Gylf 33 
Gylf 33 
Skm 16 
Gylf 33 
Skm 16 
Gylf 33 
Skm 16 
Gylf 41 
Gylf 32 
Skm 16 
Skm 16 

FFC 110 Chapter IX, Loki as the enemy of the gods 189 

In accordance with what might he expected, the Snorra 
Bdda presents a summary of all the examples found both in 
Eddie and skaldic poetry, only with the exception of three 
cases, where Snorri is our only authority viz for his mother 
Nal, his brother Helblindi anf his son Ali. The last one 
assuredly is a mistake, as Ali or Vali is elsewhere the son 
of Othin; Helblindi is not certain either, as we find it in 
(irimnismal st. 46 as a name for Othin. Nal may be a synony¬ 
mous expression for Laufey. 

But the other kennings are not equally divided between 
the skaldic and the Eddie poems. In both we find mention 
of Bfleistr, the Fenrisulfr, the MiQgarSsormr and Narfi. 
Exclusively Eddie are Laufey, AngrboQa and the birth of 
Sleipnir; only skaldic Farbauti, Sigyn and Hel. Among these 
kennings we may distinguish several groups: 

1. Those belonging to his punishment: Sigyn, Narfi, 

2. Demoniac beings: AngrboSa, Hel, Fernrisulfr, Mi5- 

3. Those showing some connection with Othin: Sleipnir, 
Ali, Helblindi, 

4. Quite unknown names: Farbauti, Laufey, Nal, By- 

We observe that the ascending line of Loki’s family 
presents a series of names, that are perfectly obscure to us, 
although etymology has tried to shed some faint light on 
them. On the contrary the descending line consists of beings, 
that appear in other myths as well. We might explain the 
first peculiarity by assuming that these names belong to a 
very old stage of religions development, whence only obsolete 
names have come down to us (cf. also the cases of Ullr and 
Heimdallr); on the other hand it is not less possible, that they 
are mere inventions of poets and mythographers. The 
examples of the second kind may help us in determining the 
character of Loki in these connections. 


Jan de Vries, The problem of Loki 

FFC 110 

If we are able to point out a group of these mysterious 
beings in other connections as well, we might expect to 
find a more solid base for the distinction of the different 
elements, that have undoubtedly been mixed up in course 
of time. The Finnish scholar E. N. Setala has tried to do so 
in a paper on »Louhi und ihre Verwandten)). 1 He studies a 
Finnish magical song about the origin of diseases, in which 
we find an old hag, called Louhi and Loviatar (and with 
many other slightly diverging names) who has been made 
pregnant by the wind. Then she gives birth to different 
beings, not only animals as the wolf, the dog, the serpent 2 
but also sons, representing diseases: 

Nimitteli poikiahan. 

Ruho yksi, rampa toinen, 

Kolmansi perisokea 3 

Ruho means »a cripple», rampa »a lame man», perisokea »quite 
blind». Now this is according to Setala a somewhat distorted, 
but still perfectly clear rendering of the Scandinavian tradi¬ 
tions about Loki. He is called the father of the Fenrisulfr 
and the Jormungandr; the children of Loviatar are a wolf 
and a serpent. But even more than that: perisokea is a 
clear equivalent of Helblindi . 4 Both Ruho and Rampa may 
be identified with Btfleistr, if we accept Setala’s etymology 
of this name as billeistr »who misses a foot». The wind that 
impregnates Loveatar is then the same as the Farbauti of 
skaldic tradition, which means »he who strikes vehemently)); 
this is, according to Bugge 5 a circumlocution of the storm- 

1 Finnisch-ugrische Forschungen XII, p. 210—264. 

2 We find even salon karhu , talvikko , sisalisko and satnmakko! 

3 K. Krohn , Suomalaiset Syntyloitsut p. 149. 

4 P. 222: In der Edda-mythologie erscheint ja Helblindi und 
perisokea ist eine so genaue Entsprechung dieses Namens, dass hier 
unmoglich ein Zufall vorliegen kann; cf. also p. 225—226. 

5 Studier I, p. 76. 

FFC 110 Chapter IX, Loki as the enemy of the gods 10L 

wind. Furthermore Laufey may be compared with Loveatar 
and besides this name we find in variants the word Aimatar, 
which is again the same as the Icelandic Nal, the word aima 
meaning »pine-needle». 1 

The strength of this hypothesis lies mainly in the pro¬ 
bability of the proposed etymological explanations. But 
this is, of course, a very frail base. If, however, the combina¬ 
tions of Setala are acceptable, then the connection between 
the Finnish magic song and the Loki-myth must lead us to 
the conclusion that there has existed an original Scandina¬ 
vian myth, were 1. Laufey had three sons, 2. one of the sons 
was exceedingly bad and 3. the mother at any rate by 
means of her son, gives birth to obnoxious animals. 2 3 

There are, of course, many difficulties to surmount and 
even many differences to ignore, before we can accept this 
hypothesis. Kaarle Krohn has seriously weakened it, by 
pointing out that the words Aimatar, Alcadtar and Naata 
can not be combined with Loveatar? Still less the children 
Ruho, Rampa and Perisolcea, who belong to the magic song 
of the elfshot; here they occur only in the tradition of the 
Eastern part of Finland and are obviously a later addition 
to this magic formula; in the tradition of Nyland and Savo- 
lax the three brothers are not quite unknown, but then they 
belong to other magic songs, f. i. against an abscess. They 
are, according to Krohn 4 , names for the devil in Catholic 
legends, who is both lame and blind. Bfleistr can not be 

1 Other names might be explained in the same way: Akaatar by 
the word a'as »prickle» and Naata as an inexact rendering of the word 

2 L. c. p. 237. 

3 In his Skandinavisk Mytologi p. 161 K. Krohn repeats the same 
opinion. Loveatar belongs, as Setala had already remarked in a 
previous study to the word lovi »ecstasy». Aimatar is a combination 
of Aijotar »the daughter of the wicked one» and Aijon aima »the 
needle of the devil». 

4 L. c. p. 162. 


FFC 110 

Jan de Vries, The problem of Loki 

brought into relation with the words for lame or cripple, 
but Helblindi may be, as well as Perisokea, a hypostasis of 
the devil, who reappears as the blind Ho dr in the Scandina¬ 
vian Baldr-myth. To the same representation belongs the 
blind murderer of Lemminkainen. Krohn believes that 
this blind devil ultimately goes back to the blind Longinus, 
who thrust his spear into the side of Christ. 

Axel Olrik has also expressed his doubts about Setala’s 
theory. 1 The similarities between the Finnish magic song 
and the traditions about Loki are very slight; if they really 
do exist they may be explained by an old Scandinavian 
magic formula, that goes back to a primitive Scandinavian 
myth about the origin of obnoxious beings. Although Olrik 
is very careful not to enter into the details of the relation, 
he feels himself justified to prove by means of this magic 

1. that the Loke of popular tradition is quite indepen¬ 
dant from the god of the Bdda, 

2. that Loki as the representative of the principle of 
evil is the last stage of the development, because he 
does not play the principal role in this magic song. 

3. that the relatives of Loki can not be divided into two 
groups, one of the principle of evil and another of 
cosmic phenomena, because the Finnish formula 
shows that the wind is his father. 

It is rather strange, that Olrik who concedes that the 
connection between the Finnish and Scandinavian traditions 
is exceedingly insignificant, still makes use of it to prove 
that his own theory about the meaning of Loki is reinforced 
by it; only the overestimation of popular lore can account for 

1 Lukki og Loviatar i finsketrylleformlerinDanskeStudierl912 
p. 95 sqq. 

FFC MO Chapter IX, Loki as the enemy of the gods 


the importance he attributes to the magical song about 
Loveatar and Perisokea. 

I am unable to follow Olrik along this way. The late 
M. Cahen was also of the opinion, that there is no relation 
whatever between the Finnish magic formula and the myth 
of Loki 1 . I am no less convinced of the necessity to refrain 
from far-reaching conclusions about hypothetical primitive 
myths on the basis of traditions, that show not only a diffe¬ 
rent character, but belong also to different times and places, 
and that can, moreover, only be brought into relation with 
each other by means of wild etymologies and doubtful 

We must try to disentangle the Icelandic traditions. 
Such a rich and varied collection of relatives is only concei¬ 
vable as the result of a long development. How did it take 
this shape? Let us begin with Loki’s brothers; they are called 
Helblindi and Byleistr. If we explain the former name as 
»the quite blind one» (hel being a weakly stressed form of 
heil ) or as »the blind one belonging to hell», in both cases 
we can easily believe that it must denote a demoniac being. 
As Mogk expresses it 2 »wohl erst auf Island ist ihm (i.e. 
Loki) ein Bruder Helblindi geworden, die personifizierte 
Finsternis unter der Erde, die noch jetzt jeden unheimlich 
ankommt, der ohne Licht in den Lavahohlen Islands weilt». 
The same name however, is according to st. 46 of Grrimnis- 
mal an epithet of Othin and if Falk’s explanation of the 
word 3 as a mistake for Iierblindi is not quite satisfactory, 
the fact itself remains unshaken, that Helblindi in a very 
trustworthy tradition denoted Othin. Did it ever signify 
anything else? There is not the slightest shadow of proof 
for it, as our only source is the Snorra Edda, where we 

1 Revue de l’histoire des religions XGII (1925) p. 65—66. 

” Hoops’ Reallexikon der germanischen Altertumskunde III, 
p. 165. 

3 Odensheite in Skrifter Vid. Selsk. i Oslo 1924 No. 10 p. 16. 



Jan de Vries, The problem of Loki 

FFC 110 

merely get the information, that Helblindi and Byleistr are 
the brothers of Loki. Obviously, this does not imply that 
they are demons. 

About the name Byleistr many etymologies have been 
proposed. Some scholars have explained the word as an 
adaptation of a foreign word, such as Belzebub 1 or Belit. 2 
Others have brought it into connection with the lightning, 
most commonly taking the second syllable -leistr in the 
sense of leiptr. 3 We have already mentioned Setala’s opinion. 
Finally Olrik translates the word as »the stormfooted» 4 
and F. Jonsson says in a slightly different way 5 : who goes 
over the fields». And it is gain Olrik, who clearly draws 
the necessary conclusion: Byleistr is a name for Othin, the 
swiftfeeted god of the stormwind. There is no single place 
in the Old-Norse literature, that may be opposed to this 
explanation; the name is exclusively found in the fossilized 
construction Byleists brodir for Loki. I think there is noth¬ 
ing more probable than that Loki is considered to be the 
brother of Othin, when he is called the brother of Helblindi 
and Byleistr. If it is said, that people could readily arrive 
at the conclusion, that Helblindi and Byleistr were two 
new gods 6 ; we are obliged to add, that this people were 
modern scholars, for in the Old-Norse tradition there is 
nothing that points in this direction. If we remember the 
well-known line of the Lokasenna, where Loki speaks about 
Othin being his »bloodbrother», we are once more convinced 

1 S. Bugge , Studier I, p. 72 and Golther , Germanische Mythologie 
p. 411 with some hesitation. 

2 K. Krohn, Skandinavisk mytologi p. 162. 

3 Wadstein , Arkiv f. nord. Filologi XI, p. 77; Mogk , Grundriss 
III 2 , p. 348; Herrmann, Nordische Mythologie p. 406; E. H . Meyer , 
Germanische Mythologie p. 276. Yglusp4 st. 51 has Byleipts. 

4 Festskrift Feilberg p. 565. 

5 Godafrsedi p. 97: s4 sem fer yfir byggdir. 

6 Miss Gras l.c. p. 113. 

FFC 110 Chapter IX, Loki as the enemy of the gods 


that this was the original conception, underlying also the 
kennings Byleists and Helblinda brodir. 

In the old skaldic tradition Loki is the son of Farbauti, 
the brother of Bfleistr, the husband of Sigyn. Moreover 
he is the father of Hel, the Fenrisulfr and the Jormungandr; 
also of Narfi or Nari. About Farbauti we know nothing; 
the name is explained as »the oarsman*) 1 , *>the storm*) 2 , »the 
lightning*) 3 or even »the demon that sends the elfshot*). 4 
I take it with Olrik 5 , that Bugge’s conception is most 
plausible, without being assured, that it renders the original 
meaning. At any rate the son of Farbauti, the brother of 
Othin and the husband of Sigyn is not necessarily a ma¬ 
licious demon. In this sense he may be explained as the 
father of such beings as Hel, the Fenrisulfr and the world- 
serpent. But we must be careful not to admit too readily 
the later conceptions of these beings, which are manifestly 
unfavourable. The Jormungandr is, after all, part of the 
cosmic system; this serpent upholds the world, as long as 
it lies coiled round the horizon. Hence it is not exclusively 
demoniac; only when it leaves its place and rushes towards 
the abode of the gods, is it taken as an obnoxious demon. 
But this belongs to the story of the Ragnarek and may 
not be placed on the same level as the religious conceptions 
of older times.' The tale of Thor’s fishing up the world- 
serpent is a well-known legend about a sea-monster, prob¬ 
ably influenced by Christian ideas; if the poet of the Hymis- 
kviba calls it the umgiord allra landa, he makes in my 
opinion a quite arbitrary identification. 

If Othin is a god of the dead, it is but natural that Loki 
is nearly related to . Hel. The conception, that he is the 

1 Wisen , Odin och Loke p. 65. 

2 S. Bugge , Studier I, p. 76. 

3 Axel Kock , Indogerm. Forschungen X, p. 101. 

4 Celander, Lokes mytiska ursprung p. 125 note. 

5 Festskrift Feilberg p. 566. 


Jan de Vries, The problem of Loki FFC 110 

father of the Fenrisulfr may* the same religious 
sphere. I do not intend to enter into the'details of the pro¬ 
blem of Fenrir; it is almost certain that' foreign, and not¬ 
ably Christian influence is traceable in tihe development 
of this demoniac being. 1 But this influence-has been ex¬ 
erted very early, for in the end of the 10th century we find 
already the characteristic traits of Fenrir. Its frightful 
appearance in the general upheaval of ; the world is most 
vividly depicted in the Eiriksmal, where 1 we read in st. 7 
(according to J. Sahlgren’s emendation 3 ): i; - 

0 , f 

t>viat ovist ’s at vita » 

mer ulfr enn hosvi ! 

scekir a sjot go5a. •*' 

In his Hakonarmal the last Norwegian poet Eyvindr skal- 
daspillir has imitated the Eiriksmal; the same idea of the 
wolf Fenrir is used by him to denote an impossibility (st. 20): 
mun obundinn a yta sjot fenrisulfr fora, adr jafngodr a auda 
trod konungmadr komi. In one Of his lausavisur (st. 6), 
written at about the same time as, the Hakonarmal (961— 
962) he makes an allusion to the gomsparri of the wolf, by 
using as a kenning for sword: Fenris varrd sparri. 3 Although 
the poet tries to represent king Hakon as a protector of 
the heathen religion, who was received after his death in 
Walhal by Othin himself 4 , he may have’been under the 
influence of his lord, who had received a Christian educa¬ 
tion. But this can not account for these allusions in his 

1 See K. Krohn, Finnisch-ugrische Forschungen VII, p. 156—161 

and Hj. Falk in Sproglige og Historiske afhandlinger viede Sophus 
Bugges minde, p. 139—144. • ’ » 

2 Eddica et Scaldica I, p. 11. 

3 F. Jonsson, Skjaldedigtning B I, p'. 63.. 

4 F. Jonsson, Den oldnorske og oldislandske Litt^ratui'shistoric I 2 

p. 451. “ : ..• • 

FFC 110 Chapter IX, Loki as the enemy of the gods 197 

poems, because they presuppose a myth known to every¬ 
body and not? a new-fangled tradition. Hence as early as 
961 the figure of Fenrir is fully developed: he has been bound 
by the gods and he will be freed when Ragnarok draws 
near. It seems possible to me, that Fenrir originally is a 
demoniac being, expressing the idea of death; its connec¬ 
tion with.Othin is not impossible as this god is accompanied 
by the wolves Geri : and Freki. It needs no further proof, 
that this idea may be exceedingly old; we find f.i. the same 
idea in the Atharva Veda, where is said in prayer (II, 29,6): 
Protect us, o gods, that the wolf may not devour us. 1 If 
Krappe is right , in his suggestion that the myth of Tyr’s 
hand being bitten off by Fenrir goes ultimately back to 
Gallic sculptures representing an animal devouring a human 
being or at least one of its limbs 2 , this conception must 
be as old as the beginning of our era. This, however, is by 
no means certain. At any rate, we are entitled to consider 
a good part of-the traditions about Fenrir, as belonging 
to the heathen period and being the result of genuine Scandi¬ 
navian development. 3 In our oldest sources Loki is nearly 
related to such a demon of death. 

We arrive at the conclusion that Loki, according to the 
oldest kennings is closely connected with Othin and at the 
same time with several demoniac beings belonging to the 
realm of death. Here, of course, there is no contradiction. 
But we get the impression that in course of time, Othin- 
and Loki developed in two opposite directions, Othin be¬ 
coming one of the chief gods, Loki on the contrary remain- 

1 Oldenberg , Die Religion des Veda p. 539. 

2 fitudes de mythologie et de folklore germaniques p. 11—27. 

3 That is why I can not accept the etymology of Fenrisulfr as 
fen-hris-ulfr (S. N. Hagen in Maal og Minne 1910 p. 57—59), because 
this explanation starts from the idea that Fenrir is a Scandinavian 
development of Behemoth, which according to Hiob lies hidden in 
rushes and watery places. 

198 Jan de Vries, The problem of Loki FFC 110 

ing in close connection with, a couple of demons. This may 
be the reason, why he stays on a level between the gods 
and the demons, and it is only natural that under the in¬ 
fluence of the Christian ideas about the devil and his realm, 
he was very easily identified with Satan and consequently 
debased to a relentless foe of-the gods. 

Narfi belongs to the same group of death-demons. The 
kenning for Hel jodis TJlfs ok Narfa proves that the wolf 
and Narfi are closely related beings. If we -may explain 
this name with S. Bugge 1 as »the narrow one», it seems to 
be a very appropriate designation of a being belonging to 
the realm of death. 2 3 About Laufey and Nal nothing can 
be said with any certainty; the last word may be a skaldic 
invention as a counterpart of Laufey. This name, how¬ 
ever, is not less the product of later tradition, as we find 
it only in such recent poems as the Lokasenna and the 
Irymskvida. It means literally: »the island with foliage- 
trees», but has this anything to do with Farbauti, who if 
he is lightning strikes the trees, or, if he is storm, may up¬ 
root them? The line of HyndluljoS st. 48 61 ulf Loki vid 
Angrbodo belongs to a not less recent period, when mytho- 
graphers invented all kinds of allegorical beings to fill- the 
gaps of tradition. 

4. The devilish Loki 

If we do not accept any longer that Loki is only an 
abridged form of Lucifer s , we are no less convinced than 
the scholars of the mythological school of Bugge and E. H. 
Meyer , that the Christian devil has exerted a great influence 
npon the later development of the Scandinavian Loki. It 

1 Helgedigtene p. 96—97. 

2 It is also connected with the idea of night; according to the 
Gylfaginning the father of Ndtt was called Norvi Or Narfi. 

3 S. Bugge, Studier I, p. 70. 


FFC 110 Chapter IX, Loki as the enemy of the gods 

is already sufficient to look at the circumlocutions, men¬ 
tioned by Snorri in his Gylfaginning ch. 32, to become 
aware of the great distance between this Loki and the god 
of the lijazi-myth. Snorri calls him rogberi ascmna, frum- 
kvedi flcerdanna and vamm allra goda ok manna. But since 
we do not find these kennings in any skaldic or Eddie poem, 
we may safely conclude, that Snorri has coined them him¬ 

But they prove the rather unfriendly feeling of later 
Icelandic tradition towards Loki. His relationship with 
several dangerous demons as Fenrir and the worldserpent, 
his flyting of the gods, his heinous treachery against Baldr, 
his place among the demons in the Ragnarok, they all point 
in the same direction. It is, however, quite unnecessary 
to insist upon this side of Loki, because it has been done 
several times before. 1 But I wish to repeat the words, I 
have said in a previous chapter, that the influence of Cauca¬ 
sian legends about a fettered demon is in many respects 
still obscure and by no means beyond all doubt, whilst the 
Christian legends are introduced first in the time of the 
disintegration of the heathen belief. Hence an important 
part of the development of Loki must belong to the pagan 
period itself. If as early as 961 Fenrir is the fettered demon 
of Ragnarok and at the same date Loki is called his father, 
we may conclude with a fair amount of certainty, that the 
germs of this development reach back to a period several 
centuries earlier. The drawback of those theories, which 
accept in a large measure the influence of Oriental and 
Christian religious representations, lies in the danger that 
we are too readily inclined to neglect the elements lying 
in the heathen religious conceptions themselves. Loki 
could never have adopted the character of Satan, if he had 

1 F.i. by Bugge, E. H. Meyer, Golther, A. Olrik, K. Krohn, to 
mention only the best names among many others. 


Jan de Vries, The problem of Loki 

FFC 110 

not been predisposed to it. This predisposition is, however, 
of the utmost importance to him who tries to solve the 
problem of Loki. For this god is not a kind of mosaic, 
where all kinds of variegated stones have been fitted to¬ 
gether into a fanciful pattern, but he must be understood, 
as every other religious conception, by a natural growth 
from more primitive conceptions according to tendencies 
lying hidden in these selfsame original representations. 


1. L o k i’s relation to the gods 

F. Jonsson has once given the following explanation 
of Loki h »He wished to do damage to the gods and to man¬ 
kind as a revenge for the death of Ymir and for all the 
giants had had to suffer. This is best proved by his name: 
he is the being who makes an end to everything (harm er 
sa son, lykur dllu). Then all becomes quite clear. Loki 
belongs to the family of the giants. He even contrived by 
his cunning to seduce Othin to a fostbrasbralag. By his 
handsome appearance he succeeded in winning the love 
the goddesses. He is, in one word, the personification of 
all that is bad and wicked.» 

The famous Icelandic scolar is obviously disinclined to 
regard Loki as the result of historical development. He 
takes him according to the descriptions, afforded by the 
sources. Here the evil nature of this god is strongly empha¬ 
sized. Hence this must be his real character. All his deeds 
must be judged from this standpoint. That is why the blood- 
brothership with Othin is explained unfavourably: Othin 
has been the victim of a. trick, played on him by Loki. 
When he accompanies Thor, it is only to show him the 
way of perdition. Even the jrymskviSa has to serve as 
a proof for this contention: the poet has no longer under- 

1 Go6afrae5i ±\ordmanna og Islendinga p. 96. 

202 Jan de Vries. The problem of Loki FFC 110 

stood, that Loki was the actual thief of the hammer, hut 
that this must have been the original idea is shown by the 
Esthonian folktale about the instrument of the thundergod. 

The reader will not want a refutation of this very simple 
theory; all the previous chapters bring sufficiant evi¬ 
dence, that it could not be true. Still the fundamental idea 
of F. Jonsson is not altogether wrong: he has tried to find 
a conception of Loki by which all the myths about him 
could be explained, without having recourse to different 
stages of development or the coalescence of several originally 
independant beings. It is, however, a mistake to begin 
from the unfavourable conception of Loki, as we find ex¬ 
pressed in the later sources. If we are unwilling to recon¬ 
struct a history of this religious conception, we are not 
allowed to close our eyes to the undeniable contradictions 
of his character, as is described by the literary documents 
of different ages. 

I wish to repeat shortly the result of our investigation: 
Loki is the faithful companion of Thor, subordinate to this 
god, even often helping him by his clever counsels. Accord¬ 
ing to the Lokasenna he has in former times sworn blood- 
brothership with Othin and there is no reason whatever 
to suspect him of any bad intentions. In the myth of Jjjazi 
he is, together with Othin and Hcenir, exposed to the ma¬ 
lignity of the giant-bird. 

This is one side of his character. But directly opposite 
are the numerous tales, where he is the unrelenting foe of 
the gods, making mischief when he sees a proper occasion,’ 
planning even the final destruction of the gods. I am of 
opinion, that Christian influence has largely contributed 
to the development of the unfavourable character of this 
god, especially because it is found most prominently in 
later traditions. But this does not imply of course, that 
the Loki of heathen religion has been wholly exempt from 
such demoniac elements; quite to the contrary, it would 

FFC 110 Chapter X, The character of Loki 20:( 

be unintelligible, if be bad been taken as a prototype of 
Satan, when there bad been no germs for such a develop¬ 
ment. But we must be careful of introducing our ethical 
ideas into a conception, that possibly was not at all con¬ 
cerned with the ideas of good or bad. As I have said before, 
his double character must be understood from the point 
of view, that he belonged to the gods as well as to the de¬ 
mons, that he by his own character stood on the border 
of these opposite supernatural worlds. Gods themselves 
are not always of a consistent character; they may be the 
protectors of mankind and the defenders of the cosmic 
order and still there may be occasions, where they show 
themselves of an irascible nature, even inclined to play 
abominable tricks or to perform atrocious injustices. We 
do not enter into the religious feelings of the heathen be¬ 
lievers and consequently all the traditions lie on the same 
level and they are to be judged by the same measure of 
critical understanding. But in real belief it is always quite 
different. Even for the Christians the Lord is one time the 
God of love and pity, at the other time the God of revenge 
and wrath. It is of no importance, that we may distinguish 
these diverging characters by an opposition of conceptions 
belonging to the Old and the New Testament, for in the 
mind of the believer they are both present at the same time. 
According to special circumstances and especially to the 
actual state of mind, the one conception may predominate 
or the other. Hence they will only very seldom come into 
collision with each other. But the man, who stands outside, 
and who reads the traditions, emanating from these differ¬ 
ent conceptions, in the same unaffected state of mind, is 
liable to the error, that here two irreconcilable ideas are 
present at the same time, and that they must be the con¬ 
sequences of totally different origins. 

Of course, I do not pretend, that Loki ever belonged 
to the domain of real belief. There are at any rate no indica- 

204 Jan de Vries. The problem of Loki FFC 110 

tions whatever of a cult, of which he has been the centre; 
we never hear of any sacrifices made to him nor are there 
place-names, preserving the recollection of heathen venera¬ 
tion of Loki. But this a question of minor importance. 
There are more gods in the same condition as he: Heimdallr, 
Hcrnir and several female divinities. It would he a rather 
rash conclusion to say that they also had never received 
any veneration or worship. A god without a cult is only an 
abstraction, nothing else but an invention of mythological 
literature. Such may be the case of Dagr and Nott, For- 
seti or Fjorgynn, but it will not be easily accepted in the 
case of divinities, that play such a preponderant role in 
the heathen traditions as Heimdallr and Loki. The cult 
may have been of a special kind; only belonging to private 
affairs, or exerted under special circumstances; it may also 
have been subordinated to the worship of a more prominent 
deity. Here all guess-work is superfluous. Hence I do not 
wish to stress this side of the problem. But I am strongly 
convinced that a supernatural being such as Loki may have 
had originally an ambiguous character and that it is by 
such a character that the diversity of his roles in the extant 
literature can be best explained. 

2. Loki’s relation to natural phenomena 

The connection of Loki and fire has been for a long 
time one of the most certain conclusions of students in the 
field of Scandinavian mythology. I had the opportunity 
to discuss this opinion in ch. VIII; here I wish to give some 
few additions. I may adduce as a curious example the inter¬ 
pretation given by E. Mogk in the Reallexikon der germa- 
nischen Altertumskunde 1 because we find here the full 
array or arguments. Mogk mentions the following points: 

1 Yol. Ill, p. 162 sqq. 

FFC no 

Chapter X, The character of Loki 2o."> 

1. Originally he is a fire-elf (Feueralf) and as such quite 
independant. As an elf he was connected with Othin, 
as a fire-demon with Thor. 

2. Because flame is thought of in some connection with 
the spirits, the purely natural phenomenon developed 
into a spiritual being, i.e. an alf. 

3. Etymology supports this view: Loki is the same word 
as lor/i; Farbauti means »he who creates damage (or 
fire) by striking»; Laufey or »island of leaf-trees» 
means the object, from which fire emanates. Both 
names are a reminiscence of the practice of the fire- 

4. Loki is a kind of culture-hero who procures fire for 
mankind, as is shown by the Brisingamen-myth. 

5. Loki is a god of lightning, vis. Byleiptr which means 
»Donnerblitz», Sigyn that is »the rain-cloud», Loptr 
or »the lightning» and Nari a personification of the 
cold winter. 

It is hardly necessary to show the frailty of this hypo¬ 
thetical construction. It is built exclusively on the theory 
of Olrik about Loki’s nature as an ildvcette. We will have 
to discuss it presently in our chapter XI. As I have often 
said, etymology is utterly worthless; if we want to maintain 
an other explanation of Loki, the different names will lend 
themselves to it as well. The conception of Sigyn as the 
raincloud or Byleiptr as the »Donnerblitz» is a curious 
survival of old mythological theories, that explained al¬ 
most every divinity as the religious equivalent of a na¬ 
tural phenomenon. The highly constructive character of 
this theory is shown clearly by the explanation of Loki’s 
relation to Othin as well as to Thor. A fire-alf means both 
the element of fire and an elfish being; that is said to be 
the reason that Loki became attached to the thundergod 
and to the Lord of the Asir. But how this came to be, is 

-06 Jan de Vries, The problem of Loki FFG 110 

left in enigmatical darkness and it appears to me, that this 
should he of particular interest. Not every elfish being 
had the brilliant success of Loki, who became even the 
fostbroSir of Othin. Was it only on account of his elfish 
character? I am not prepared to accept it without any 
further proof, because it is by no means a development, 
implied in the original character itself of a fire-elf. Here 
is the salient point: if we can not explain the transition 
from one sphere into an other, of a simpler fire-spirit into 
the full-blown companion of the supreme gods Othin and 
Thor, then the theory is absolutely unacceptable. 

Are there indeed in the mythological traditions them¬ 
selves any indications of Loki’s fire-nature? No conclusive 
ones, at any rate. The salmon-shape of Loki can not be 
explained as a proof of it, because, as we have seen, there 
is no reason, to explain the salmon as a fire-fish. If Loki 
should be demonstrated as a fire-demon, on the basis of 
indisputable facts, then we might conclude that the salmon 
perhaps had some connection with a myth about the origin 
of fire. It would not be positively established, for the 
salmon-shape might have had an other meaning as well or 
even no particular meaning at all, particularly in such a 
literary tale as is the capture of Loki. But, certainly, some 
amount of probability for this opinion could be accepted. 
It is, however, an example of poor method, to accept the 
fire-nature of the salmon, solely relying upon the fact that 
in mythological traditions of other peoples a spark of fire 
may be concealed in a fish, and then to conclude, that Loki 
who once took the shape of a salmon, must have been a 
fire-demon too. 

There is one place in the Old-Norse literature, I have 
hitherto left out of discussion. At the end of the Lokasenna 
it is said, that Loki disappears from ACgir’s hall after having 
pronounced the following threat: 

FFC 110 Chapter X, The character of Loki 207 

eiga Jtin oil er her inni er 
leiki yfir logi 
ok hrenni ]ser a haki! 

Several scholars have contended, that it is clearly expressed 
here, that Loki has power over the element of fire . 1 For 
if he had been f.i. a waterdemon, he assuredly would have 
said, that he would destroy the gods by drowning together 
with JEgir’s hall. After all this is a very superficial con¬ 
clusion. Celander has shown by the analogy of several folk¬ 
tales, that a malevolent being may punish by setting a house 
on fire . 2 Of course, he explains Loki’s words as a proof of 
his opinion, that Loki is originally a vcette, an elfish being. 
Olrik gives his assention to Celander’s folkloristic parallels, 
but explains the words of Loki as a prediction of the Ragna- 
rok, that presently will destroy the world by fire . 3 

Perhaps the explanation is still more simple. Loki is 
in the hall of the sea-god; if he wishes to destroy it, he can 
not do it by means of water, for this will not make any 
impression upon the lord of the ocean. So he must take 
the element, opposite to iEgir’s' nature, vis. fire. But the 
author of Lokasenna has certainly felt a connection be¬ 
tween the names of Loki and the word for fire; it is a kind 
of pun, when he says in his last words to the assembled 
gods: leikr yfir logi. But this is not a connection of or¬ 
iginally equivalent ideas, but only a question of popular 
etymology. Here we find the same case as in the myth of 
Thor’s journey to Utgardaloki, where Loki and Logi are 
described as opponents. I may refer the reader to my 

1 F.e. Wisen , Oden och Loke p. 69; A. Koch , Indogerm. For- 
schungen X, p. 90 sqq.; Herrmann , Nordische Mythologie p. 409; 
E. H. Meyer , Germanische Mythologie p. 278. 

2 L.c. p. 78—80. 

3 Festskrift Feilberg p. 567 note 2; cf. also Gering-Sijmons , 
Edda Kommentar ad Lokasenna 65. 

Jan de Vries, The problem of Loki 

FFG 110 


discussion of this tale in ch. V, 3. It is a very self-evident 
result of the highly intellectual activity of the Icelandic 
mythographers, who treated the old religious lore as a 
purely literary tradition and were not at all adverse to 
allegorical interpretations. No one can deny the fact, that 
Loki actually in a certain period of Old-Norse tradition, 
has been connected with the idea of logi, but there are, as 
far as I see, no proofs whatever that the original meaning 
of Loki has anything in common with fire. 1 

Then why not try the element of w a t e r? In my 
second chapter I have mentioned the view of Wisen, who 
indeed accepted Loki as being connected with this element 
and who adduced as an argument in its favor an etymology 
of Fenrir. Axel Olrik has collected all the material, that 

1 As an amusing example of learned explanations about Loki 
without any real foundation in the actual traditions, I wish to quote 
the following sentences from L. von Schroeder , Mysterium und Mimus 
im Rigveda p. 219: »Loki ist das Feuer im Ofen, ganz elementar 
und damonisch gefasst, das Feuer als gefahrliches Element, das 
man hiiten und besanftigen muss. Loki ist auch die Sonne, sofern 
sie gltiht und brennt, durch ihre Hitze versengt, Wasser zieht und 
Diinste entwickelt — kurzum die Sonne von ihrer elementaren 
Seite genommen, insbesondere auch sofern sie qualt und lastig wird, 
Vegetation und Menschen schadigt. Das Brfsingamen dagegen ist 
die Sonne, sofern sie als herrliches Kleinod der himmlischen Gottin 
da droben strahlt, Gotter und Menschen durch ihre hehre Schonheit 
entztickt; und es ist zugleich das Feuer, wie es beim Sonnwendfest 
lodert, das Feuer, sofern es als hehres und heiliges Symbol die Sonne 
darstellt, die Sonne griisst und in ihrem Aufstieg starkt. Darum 
steht in gewisser Weise das Brisingamen den heiligen Gottern Agni 
und Apollon seinem Wesen nach naher als der tiirkische schaden- 
frohe Loki.» It will hardly be necessary to criticize these nicely 
constructed sentences, where only the words »hehr» and »heilig» are 
somewhat too frequently used. The fire on the hearth is, to begin 
with, not demoniac and not dangerous, but on the contrary a benefit 
to mankind. And what on earth Loki might have in common with 
a fiercely burning sun (perhaps a kind of demon meridianus?) is to 
me quite unintelligible. 

Chapter X, The character of Loki 

FFC 110 


could point in this direction and he draws our attention 
to the following facts: 1. the Andvari-tale, where Loki 
kills an otter and catches the pike in the waterfall; 2. the 
invention of the fishing-net; 3. his salmon-shape to escape 
from the persecuting gods; 4. his seal-shape in the fight 
with Heimdall; 5. the Faroese folktale where he is lord of 
the fishes. 1 Still Olrik does not venture to draw the con¬ 
clusion. that Loki could have been a water-god too and this 
is. indeed, the only possible view. After the treatment of 
the myths and of the Faroese folktale, where Loki may 
seem to have some aquatic proclivities, in our preceding 
chapters, it Can not but be evident, that they are quite too 
shaky a foundation to build such a theory on it. Loki is 
no more a god of the water because he takes the shape of 
a fish, than he is a god of the air on account of his flying 
in Freyja’s feather-apparel or of his being occasionally an 
insect. It is of no importance whatever, if the Flemish 
spirit Kludde also shows several traits, that may be ex¬ 
plained as indications of a special connection with water; 
Miss Gras 2 , who stresses this argument, forgets that the 
basis of the identification of Loki and Lodder-Kludde is 
only the alleged etymological correspondance of the words 
Lodder and Loburr, and that in the image of the latter, as 
we know from Scandinavian tradition, no single trait points 
to the element of water. Moreover, if we could indeed 
assume, on account of a couple of clear and incontestable 
examples, that Loki in some cases had the character of a 
water-god, it would be impossible to explain his complicated 
personality from this standpoint alone and we should be 
obliged to seek for other original traits, by which the remain¬ 
ing myths — and to be sure the majority of them — could 
be understood. 

1 Festskrift Feilberg p. 569. 

2 L.c. p. 123. 



Jan de Vries, The problem of Loki 

FFC 110 

Gods, that represent a definite part of nature, are now- 
a-days not hailed with such confidence as a quarter of a 
century ago. We have come to the insight, that in the 
framing of the supernatural world man wanted to create 
a transcendental correspondence to his own needs and his 
own interests. Gods are not the result of the contemplation 
of the universe, hut of the keen desire to have companions 
and helpers of a higher order in the troubles and labors of 
earthly life. As far as this life is dependant upon nature 
and the manifestation of its forces, representations of this 
kind may enter into religious conceptions. If people do 
believe in a fire-god, it is only because the necessity and 
the usefulness of fire has in all ages been present in the mind 
of the worshipper. It was the act of making fire, the means 
to use and to control it, the danger emanating from it, that 
compelled mankind to subject himself to rules and actions, 
that might have the desired effect. But his attitude towards 
fire, although eminently practical as to its purpose, was of 
course quite intermingled with magical practices and re¬ 
ligious conceptions. This makes it often extremely difficult, 
especially in a religion that has become more or less sophist¬ 
icated, to distinguish between a belief in fire as a natural 
element and the religious attitude towards one of the most 
valuable necessities of human life. Fire, however, is in 
most cases not a subject, but purely an object. It must 
be acquired, and when it is kindled on the hearth it must 
be treated with the utmost care. Fire in itself is not a 
divinity, but it might be easily conceived of as having a 
divine origin. 

3. Loki’s relation to the underworld 

In his paper on the Teutonic heaven-god, Ii. Much 1 
considers the evidence of the etymology of Loki’s name as 

1 Der germanische Himmelsgott p. 57. 

FFC 110 Chapter X, The character of Loki 211 

belonging to the verb lukan »to close» and thinks it possible 
that this word, to which ags loca »prison» belongs, might 
have originally denoted a subterranean place of punish¬ 
ment, being a distinct part of the realm of death; Loki 
might have developed into its ruler as the first and most 
important of those who had been hurled into it. This Loki 
is of course only Lucifer in Scandinavian disguise; if Much 
contends, that this could have taken place by »an inde¬ 
pendant development of the Teutonic myth», it is not easy 
to follow his idea and it does not alter the fact, that such 
a Loki is a foreign element in the religious system of the 

Indications for the chthonic character of Loki, although 
very scarce, are not wanting. In the Fjolsvinnsmal the 
weapon by which the wondrous bird Vidofnir may be killed 
is called Lcevateinn, and in st. 26 it is said: 

Lievateinn hann heitir en hann gorbi Loptr runom 

fyr nagrindr neban; 

i Sfegiarns keri liggr hann hia Sinmoro 

ok halda niarblasar nio. 

There are nearly as many riddles in this stanza as there 
are proper names; we do not know the meaning of the 
Lfevateinn, nor of Sinmara and the box of Snsgiarn. They 
may be poetical embellishments of later ages. But it is 
clear, that Loptr did make this fearful weapon, that he did 
it by means of magical runes, while he was within the 
enclosure of the Underworld. It is, perhaps, not surprising 
that Loptr or Loki is supposed to be in the realm of death, 
but it is, on the contrary, a rather unexpected statement 
that he forges a weapon and makes use of runes. Celander 
is right, when he says that this tallies excellently with the 
character of an elfish being, although it does not follow 
that it proves Loki’s character as such. 1 Loptr is conceived 

1 L.c. p. 120. 


Jan de Vries, The problem of Loki 

FFG 110 

of as a dwarf in this stanza and we may even say that the 
poet of this part of the Svipdagsmal indeed had this view 
of the god, for in st. 34 he mentions Loki as the builder of 
the palace of MengloS together with incontestably dwar¬ 
fish beings such as Dellingr, Ori and Varr. 1 2 So we may ask 
if this represents an original tradition; a line as e.g. Uni ok 
In, Ori ok Bari makes the impression of a soundplay, void 
of any particular mythological meaning. Such lists of 
names as we find them in the Snorra Edda contain only 
partly genuine material of old lore; the inventive mind of 
poets and mythographers has surely added much to them. 
Loki, of whom the Soria fattr says that he was ekkimikill 
voxtum has been brought into connection with elfish beings, 
although he does not originally belong to them. But he 
was a clever spirit, prone to mischief and velrcedi; he stood 
in opposition to the gods as well as to the giants; this may 
have been sufficient to compare him with dwarfs and other 
elfish beings. The characteristic of the Sorlajiattr may be 
considered as the result of this later development and it 
is at any rate no proof for the high antiquity of this concep¬ 
tion. But it lay ready to hand; if we only think of the slow- 
witted Thor with his huge body and tremendous force we 
can easily understand that the shrewd Loki, who was in 
every way the opposite of the thundergod, was represented 
as a great intellect in a nimble body. 

The stanza of the Fjolsvinnsmal does not yield such 
strong evidence for Loki’s character as a death-demon, as 
Schoning 2 would have us believe. Its content is highly 
suspicious, and moreover, as v. Sydow objected in his paper 
on the giants 3 , the poem says only that Loki made this 
weapon on this subterranean spot, but not that he was in 

1 These names are found in the VQluspd and the I>ulur of the 
Snorra Edda. 

2 Dodsriger i nordisk hedentro p. 40. 

3 Jattama p. 16. 

FFC 110 

Chapter X, The character of Loki 


the habit of living there. Loptr is said to have done this 
fyr nagrindr nedan, because this would aptly give us the 
impression of something very horrible and terrifying. 

The same may be said about stanza 23 of the Lokasenna, 
where Othin says to Loki: 

atta vetr vartu fyr ior6 ne5an, 
kyr molkandi ok kona. 

For if we accept this as being an allusion to a genuine myth, 
then we know nothing more about it than this couple of 
lines and we are unable to determine the character of Loki, 
when he, as a woman, has been milking cows 1 in the under¬ 
world. In this case he can hardly have been considered as 
a chthonic being, but he committed an infamous act by 
taking the shape of a woman and doing the work of a milk¬ 
maid. Was it on this account that he hid himself under 
the earth? 

It is, however, not to be denied, that heathen mythology 
has felt a connection between Loki and the world of the 
Dead. We have already seen that Hel is called his daughter 
and this is only possible if he was considered to be a chthonic 
being. Perhaps the same may be said on account of TJtgarda- 
loki, that is, the Loki belonging to Utgarb, »the world out¬ 
side the regions inhabited by mam. But it is not clear what 
the exact relation between Loki and UtgarQaloki may have 
been. There are several possibilities. Utgarbaloki may 
be an invention of later times and his name may point at 
an opposition to the real Loki. UtgarSaloki belongs to the 
world of trolls and monsters, not to the realm of death. 
Finally Loki may be a shortened form of Utgardaloki, and 

1 I accept the explanation of kijr molkandi as »milking (ows» 
and not as »a milkgiving cow» (cf. Gering-Sijmons, Edda-Commentar 
I, p. 288), in the same way as we find in the first lay of Helgi Hun- 
dingsbani st. 43: pa er pu Gullnis geitr mdlka&ir (cf. E. .1. Kock, 
NN § 10). 

- li Jan de Vries, The problem of Loki FFC- 110 

so have been originally identical. It seems difficult to me 
to choose between these conceptions without making a 
quite arbitrary decision. I am inclined to consider Utgarda- 
loki as the name of Loki in his special function of lord of 
the demoniac world, but at the same time I am convinced 
of this being the result of a later development under the 
influence of Christian ideas about hell as a place of punish¬ 
ment and torture. The Old Teutonic belief had apparently 
other ideas about it and the goddess Hel herself is only 
a personification of the subterranean cave, where the dead 
came together when they had to troda helveg. 

From all this it is obvious, that in the conception of 
Loki there must have been elements that could lead to such 
a development. His intimate relation to Othin, as it is 
expressed by their fostbrcedralag, may be explained by the 
supposition, that they both belonged to the underworld. 
But then we can not follow the argument of Hilding Celan- 
der, who considers the connection of Loki and Hel as a 
consequence of his being a spirit, which the Old-Norse 
sources usually call alf. Here again the modern representa¬ 
tions in popular lore are confused with the heathen con¬ 
ceptions; the alf in Old-Norse sources is not the teasing 
spirit of modern folk-tales, nor the wily smith of mediaeval 
romance, but a spirit of the dead intimately connected with 
the fertilizing forces at work in the bowels of the earth. 1 
It is a far cry from a benevolent halfgod like Olafr G-eirsta- 
5aalfr or BarSr Sngefellsass to a trickster as Loki appears 
to have been even in the oldest sources. 

We arrive at a conclusion, that is only provisional. 
There is no reason to deny the possibility of Loki’s connec¬ 
tion with the world of the dead, but the meagre evidence 
we can gather from Scandinavian literature does not allow 
us to make out of what character this relation has been. 

1 Cf. my paper »Van Alven en Elfen» in the (Dutch) Tijdschrift 
voor Volkskunde XXXVI (1931) p. 3—30. 

FFC 110 

Chapter X, The character of Loki 


4. Loki’s bisexual character 

In the stanza of Lokasenna quoted in the previous 
paragraph we find two more lines, that however have been 
rejected as a later addition by several scholars. They are: 

ok he fir pu par born borit 
ok hug5a ek pat args a9al. 

It may be, that these lines do not originally belong to the 
poem, but their contents are quite in accordance with the 
opinion of the author of the Lokasenna, for in st. 33 Njordr 
repeats the same insult to Loki: 

hitt er undr, er ass ragr er her inn of kominn 
ok hefir sa born of borit. 


The outrageous insult of ergi is furthermore borne out by 
the myth of Sleipnir’s birth, as it is told in the story of the 
giant-builder. Also the HyndluljoS says in st. 40: en Sleipni 
gat vid Svadilfara. A dim recollection of this myth is still 
preserved in the Faroese popular tradition according to 
which Loki himself avowed that he underwent different 
transformations, but that the worst experience, he ever 
had, had been, when he as a mare was pregnant by G-rani. 1 

It is of importance that the same is related about Othin. 
Loki answers him in st. 24 of the Lokasenna by throwing 
back at him the same outrage: 

En pik si9a ko9o Samseyjo i 

ok draptu a vett sem volur 
vitka liki fortu verjddd yfir 

ok hugda ek pat args adal. 

Othin visiting the people in the shape of a witch and per- 

1 H. C. Lyngbye , Faeroiske Qvseder om Sigurd Fofnersbane og 
hans aet (Randers 1822) p. 21 note; communication of Pastor Schroter. 


Jan de Vries, The problem of Loki 

FFC 110 

forming ignominious sorcery is scarcely any better than 
Loki and. perhaps we had better listen to the advice of Prigg, 
who says in the following stanza: 

orlogom ykkrom skylit aldregi 
segia seggiom fra, 

hvat it a* sir tveir drygQoS i ardaga; 
firriz a forn rok firar. 

But it goes without saying that these motives are too im¬ 
portant to pass over in silence. The transformation of these 
gods into women is an undeniable fact and in itself it seems 
not to be incompatible with the respect due to a deity, as 
Othin himself is not exempt from this evil. Still it is not 
a deed, that brings much honour to a man and as soon as 
there is any tendency to suspect a god of heinous crimes, 
such a tradition is apt to be explained in malam partem. 

As Axel Olrik already pointed out 1 , the most important 
among these myths, the tale about the birth of Sleipnir, 
does not necessarily present Loki as a wicked god, for he 
is here the shrewd deceiver of the troll. But as I have tried 
to prove, this myth may have been originally quite inde¬ 
pendant from the tale of the giant-builder and so we have 
only to reckon with the bare fact, that there has been a 
myth, where it was related, that Loki had given birth to 
Sleipnir and that its father was called Sva&ilfari. Whether, 
however, this myth can claim to be originally heathen, is 
open to serious doubt. 2 Kaarle Krohn is inclined 3 to admit 
Christian influence, and he draws our attention to a Finnish 
tale about Lemminkainen, who treats a blind person with 
contempt, because he has had illicit intercourse with a 
mare, and to the popular belief, that the horse is closely 

1 Festskrift Feilberg p. 559. 

2 F. Jonsson, GoSafraeSi p. 91 calls it of later origin. 

3 Ubersieht liber einige Resultate der Marchenforschung, FF- 
Comm. No. 96, p. 120—121. 

FFC 110 

Chapter X, The character of Loki 


connected with the devil. But these parallels are in them¬ 
selves not enough to prove, that the Sleipnir-myth must 
be an adaptation of a Christian legend. The abundance of 
proper names does not prove the contrary, because later 
mythographers are very liberal in inventing them. But if 
Sleipnir as the horse of Othin originally means the death- 
horse, it is by no means incredible that a myth has existed 
about its birth and then we are not entitled to exclude the 
probability, that Loki has played a r61e in it. Together 
with the other examples of his ergi, the myth of the birth 
of Sleipnir may contain a germ of truth about Loki’s char¬ 
acter, which may be useful for our investigation. 

We may finally add a few other stories about the female 
nature of Loki. In the myth of Baldr he acts twice as a 
woman, first when he asks Freyja about the secret of Baldr’s 
invulnerability and secondly when he in the shape of {jokk 
prevents Baldr from being released from Hel’s domain. 
They show us Loki in the same kind of situation, only less 
brutal, than where he is pictured as a pregnant woman. 
This is again the case in a very obscure stanza of the Hynd- 
luljo5 (41): 

Loki at hiarta 
fann hann halfsviQinn 
var5 Loptr kvibugr 
[aQan er a foldo 

lindi brendo 1 , 
hugstein kono; 
af kono illri, 
flag& hvert komit. 

The stanza is relatively late with regard to the development 
of Loki’s character; he is here the pro creator of all wicked 
beings and has been abased into a kind of heathen devil. 
How little the author, however, understood the real mean¬ 
ing of the mythological conceptions he mentions in this 

1 The first line seems to be bungled in some way or another; 
H. Pipping , Studier i Nordisk Filologi XVIII, 4 p. 46—52 suggests 
the following emendation Iokz af hiarta ividu brendu, which is indeed 
a very ingenious guess, but one that cannot be proved. 


Jan de Vries, The problem of Loki 

FFC 110 

part of the HyndluljdS, becomes clear from the preceding 
stanza where the giving birth to Fenrir and to Sleipnir are 
represented as equally bad. But although the stanza in its 
present form may be of a rather doubtful character, the 
myth itself alluded to can not be discarded in such an easy 
way. The Norwegian physician Reichborn-Kjennerud says 
about this story that it is unparalleled in Old-Norse tradi¬ 
tions 1 , but this does not imply, that it could not be of 
genuine Scandinavian origin. Axel Olrik has quoted a 
Lithuanian folk-tale, where a girl eats the heart of a burnt 
hermit and two hours afterwards gives birth to a marvel¬ 
lous boy who does not wait a moment before asking his 
grandfather to put the horses to the carriage. 2 Olrik thinks 
that the likeness between this folk-tale and the Hyndluljod is 
so great —the heart is preserved although the corpse is 
burnt; when eaten it causes pregnancy; the child that is 
born shows the same qualities as those of the owner of the 
heart —that he feels himself obliged to suppose some con¬ 
nection between them; the story in the Eddie poem loses 
all its mysteriousness — but at the same time also all its 
interest — if we consider it as an adaptation of a popular 
tale from Eastern Europe. There is no proof whatever for 
such a supposition. On the contrary, the fact that in the 
Lithuanian tale the heart is swallowed by a girl and in the 
Hyndluljod by Loki shows, that there are strong differences 
between these stories as to their meaning, and the fact itself 
that a heart is eaten and that this causes a child to be born 
belongs to a belief, that is wide spread and may be found 
in differents parts of the world. The heart is the seat of 
life and whosoever wants to obtain the qualities of a certain 

1 Var gamle trolldomsmedisin (Skrifter utgitt av Det norske 
Videnskaps-Akademi i Oslo, Hist.-Filos. Klasse 1927 No. 6) p. 24. 

2 Festskrift Feilberg p. 556 ff.; Miss Gras l.c. p. 119 repeats the 
arguments of Olrik, only adding a few very superficial remarks 
intended to throw doubt on these traditions. 

FFC 110 

Chapter X, The character of Loki 


being can not do it in a more efficacious manner than by 
eating this organ. So did Sigurd, who after eating the 
dragon’s heart obtained the knowledge of the birds’ lan¬ 
guage. The belief, that one may become pregnant by eating 
an object, a fruit, a part of an animal, or whatever it may 
be, need not be illustrated by examples as almost every 
collection of folk-tales contains many of them. In an extra¬ 
ordinarily great number of these stories pregnancy is caused 
by the eating of fish x ; a very interesting specimen is the 
irish story about Tuan mac Cairill who had been changed 
into a deer, a wild boar, a vulture and a fish; in this last 
shape he was eaten by a woman and reborn as a child. 1 2 
Hence it is continuing in the same train of thought, when 
the swallowing of a vital organ like the heart has the same 
effect. The Lapp sorcerer who, according to a Norwegian 
tale 3 , wished to eat the hearts of nine men in order to 
obtain everlasting life and the power of making himself 
invisible, obviously had the same idea about the qualities 
Which may be acquired by eating this organ. It is a world¬ 
wide belief that by eating the heart or any other vital organ 
of a human being or of an animal, that the qualities in¬ 
herent in it, are transmitted; therefore it is unwise to eat 
the heart of a hare, whilst it is highly beneficial to swallow 
the same organ of a slain enemy, a lion or a bear. 

But if this be granted, still the fact remains that the 
heart of the witch is half-burnt, which shows a remarkable 
likeness to the Lithuanian folk-tale. Still I do not believe 
that this gives us the right to derive the Old-Norse myth 

1 Cf. Bachtold-Staubli, Handworterbuch des deutschen Aber- 
glaubens II, col. 1032; Bolte-Polivka, Anmerkungen I, p. 544—545. 

2 D ’ Arbois de Jubainville, Le cycle mythologique irlandais 
p. 57. Tales about women who become pregnant by swallowing 
worms in a bucket with water, were also current in Ireland; cf. 
MacCulloch, Celtic Mythology p. 140. 

3 Cf. Reichborn-Kjennerud 1. c. 


Jan de Vries, The problem of Loki 

FFC 110 

from a popular tradition in Eastern-Europe. Witches and 
other obnoxious beings had to be destroyed by fire, if man 
wanted to get rid of them; this is the method always applied 
to the destruction of a draugr. The idea might easily arise 
that the heart, as the principal organ and the seat of psy¬ 
chical qualities, was difficult to annihilate, when its owner 
was of an exceptionally mischievous character. The poet 
calls the witch’s heart hugsteinn and it is possible, that he 
wanted to express by this kenning, that it was so hard that 
even fire could not damage it. Is it then necessary to fetch 
a modern tale from a Baltic people, only to explain a motive 
that could so easily arise on the basis of the Scandinavian 
belief itself? Again we state, that it is not enough, if one 
wishes to accept such connections between two traditions, 
belonging to different times and different peoples, when 
a few minor traits show some resemblance, more especially 
if they belong to the belief current among several peoples 
of the earth. 

But although I am of opinion, that the stanza of the 
HyndluljoS may ultimately go back to Scandinavian popular 
belief, I do not wish to conclude, that it is remnant of a 
genuine heathen myth. We can neither affirm nor deny 
it; perhaps it is safest to exclude it from our argument and 
to consider it as a pure fiction, giving fresh evidence for 
the evil character of Loki. 

The transformation of Loki into a woman and his be¬ 
coming pregnant may be considered as belonging to an 
early stage of the Scandinavian religion, although it may 
have been ultimately developed into opprobrious tales about 
this highly compromised god. The fact, that the same is 
told about Othin, makes it plausible, that we are touching 
here a very old section of religious belief, where the connec¬ 
tion between Othin and Loki was very intimate. 

Two solutions of the problem may be proposed. Olrik 
has already compared the Tacitean account about Tuisto, 

FFC 110 Chapter X, The character of Loki 221 

the Teutonic creator of mankind and it is generally accepted 
that the name of this primeval being denotes his bisexual 
character. 1 We find the same conception in the Old-Norse 
myth about Ymir, told by Snorri in his G-ylfaginning; while 
this being was asleep from the sweat under his right arm 
there came forth the first human couple and by rubbing 
his legs the hrim] ursar were born. The Indian Yama, the 
Iranian Yima and Zrvan are likewise hermaphroditic and 
several more examples might be adduced from other re¬ 
ligions as well. Therefore it becomes possible to assign Loki 
to this group of gods, but there is one serious difficulty: 
the Scandinavian religion gives to this double-sexed prime¬ 
val being the name Ymir and that this conception reaches 
back into the mists of farthest antiquity is proved by the 
undeniable connection between the words Ymir and Yama. 
Is it then possible to assign Loki a place in this selfsame 
character at the side of Ymir? 

But there are more instances of bisexual divine beings, 
notably among the gods of fertility. In Cyprus there was 
a statue of the Venus barbata, called Aphroditon by Aristo¬ 
phanes and we hear about a deity Luna to whom sacrifices 
were made by men in the clothes of women and by women 
in those of men. We are likewise reminded of the Oschophoria 
at Athens, where chosen boys were dressed up as girls and 
taught to imitate the gait and voice and bearing of maidens. 2 3 
In India we meat the divine couple of Siva and Kali, some¬ 
times represented as one single being, as an ardhanarisvara 2 
These conceptions wm e not at all unknown among the Ger¬ 
mans: Tacitus tells of a cult of the twin-brothers the Alcis 
among the Naharnavali, an East-Germanic tribe; the rites 
were performed by a priest muliebri ornatu. The Scandina¬ 
vian god Nerthus may have shown the same character, since 

1 K. Miillenkoff, Deutsche Altertumskunde IV, p. 113. 

2 Farnell, The Cults of the Greek States II, p. 628 sqq. 

3 Sten Konow in the Festskrift Kjser p. 58. 


Jan de Vries, The problem of Loki 

FFC 110 

among Germanic tribes in the Cimbrian peninsula, as Tacitus 
affirms, a goddess Nerthus was venerated. 1 There is no 
reason to doubt, that the Germans knew the conception 
of a fertility-deity in the double aspect of god and goddess, 
and when Jungner affirms 2 that among the Scandinavians 
a hermaphroditical deity could not have played any con¬ 
siderable r61e in relatively recent times, because in the 
historical sources ergi is one of the most heinous crimes, 
he does not take sufficiently into account that Saxo Gram¬ 
maticus still mentions with horror the sacrifices at Upsala, 
renowned for their »effeminatos corporum motus scenicos- 
que mimorum plausus ac mollia nolarum crepitacula». 3 More¬ 
over in religious practices of several peoples things are 
allowed and even highly recommended, that are utterly 
forbidden as moral depravations in ordinary times. The 
disgust for a man’s ergi does by no means exclude the 
possibility, that in the cult of a fertility-god the changing 
of sex may not have taken place. The idea itself of a divine 
being, having at the same time a male and a female sex is 
no proof of depravation; earth herself is in spring female, 
but in automn male, as Andre Jolles has observed in a very 
interesting paper about the changing of sex 4 5 ; what then 
is in more conformance with this natural phenomenon than 
a vegetation-god showing the same duality of character? 

May we then rally to the thesis of F. R. Schroder 8 , that 

1 For this female Nerthus I may refer to my remarks in my 
paper FFComm. No. 94, p. 37 and note 84, where I have emphasized 
the fact that the name Nerthus-NjgrSr denotes the male god, but 
this does not exclude him from being bisexual, as well as the Venus 
barbata who bears a female name although she is none the less a 
male god. 

2 Gudinnan Frigg och Als harad p. 234. 

3 Lib. VI, edition Olrik-Rader I, p. 154. 

4 Geschlechtswechsel in Literatur und Volkskunde, Schriften- 
reihe der Societat »Jocosia» zu Leipzig, Heft 5, p. 160. 

5 Germanentum und Hellenismus p. 115 sq. 

FFC 110 

Chapter X, The character of Loki 


Loki really is a vegetation-deity and even originally of a 
bisexual character? It seems to me, that in this form, the 
solution is too categorical; we must leave the possibility 
open, that Loki at some time of the historical development 
of the Scandinavian religious conceptions has shown part¬ 
icular traits that remind us of fertility-demons, but we, 
would be too rash in our conclusion, if we proceeded so far 
as to say that his original character was that of a herma¬ 
phroditic vegetation-deity. 

5. The ethical character of Loki 

Although the family of the Old-Norse gods is not alto¬ 
gether faultless as to its morals, Loki makes on the whole 
a very unfavourable impression, by which he seems to be 
in a somewhat isolated position among the deities. It is, 
of course, self-evident, that the devilish character of Loki 
has been developed under the influence of Christian ideas; 
here the conception of a principle of wickedness, of a per¬ 
petual foe to mankind and an opponent of God himself, 
was so prevalent, that it could not but attract a figure such 
as Loki about whom evil rumours were whispered. But 
even in a time, when Christian influence can have been only 
very slight, Loki stood in a close connection with such evil- 
renowned beings as He] or Fenrir. 

The myths represent him mostly as a mischief-maker. 
I have tried to demonstrate in § VII, 2, that his remarkably 
wavering position on the verge of the divine and the de¬ 
moniac worlds, does not place him in direct opposition to 
the Asir. In fact, in a great many myths he is a companion 
of the gods who is able to entice them into a dangerous 
conflict with the trolls, but who is unvariably obliged to 
mend the evil he has done. The kind of evil differs. As one 
of the chief opponents in the Ragnarok he shows himself 
in his worst character; and when lie delivers I5unn with 

224 Jan de Vries, The problem of Loki FFG 110 

her apples of rejuvenation into the power of a giant, he 
causes a dire calamity for the gods. But on the other hand, 
cutting the hair of Sif, deceiving the poor dwarf Brokkr, 
stealing a jewel from Freyja, these are more the tricks of 
a naughty boy than of a benevolent demon. 

If we call him a thief and a deceiver, we might produce 
a mass of material to prove this qualification. But I think 
it is better to give a definition as MacCulloch has done: he 
was also an embodiment of the mischief-maker, so common 
in all states of society, whose mischief has often dire results 
for himself or others. 1 I do not wish to put this as the original 
conception, from which the very complicated figure of 
Loki has been developed, but as the central point of his 
character, from which the different attitudes radiate in all 
directions. Or perhaps we had better choose an other simile, 
such as Prof. Andre Jolles suggested to me during a con¬ 
versation on this subject; the mythical tradition about 
Loki gives the impression of a funnel, into which a very 
inordinate mass of tales and ideas may be collected. Loki 
may have had a strong attractive power over a wide range 
of mythological tales and folkloristic motives, but the 
magnet that exercised this attraction, was again nothing 
else but his character of mischief-maker. 

If we wish to explain the religious phenomenon, that 
in Old-Norse tradition bears the name of Loki, we must 
start from this characteristic of mischief-maker. This can 
not be an accessory detail, but it must be the very core of 
the conception. Furthermore we have to look for an explana¬ 
tion of his bisexual character, if possible in direct connec¬ 
tion with his r61e as mischief-maker. The Old-Norse tradi¬ 
tion, however, does not give us the clue to a satisfactory 
solution of this problem; hence we have to look for more 
information in other fields of scientific research, that is, 
in folklore and in ethnology. 

1 J. A. MacCulloch, Eddie Mythology (Boston 1930) p. 149. 



1. Survey of the folkloristic material 

Axel Olrik has given a very complete collection of the 
popular traditions about Loki in a paper which I have had 
the occasion to mention already several times. 1 While I 
will refer the reader to his exposition of the extant material 
for all the necessary details, I wish only shortly to mention 
here the adduced examples in order that it may be possible 
to follow the discussion of the facts. 

I. Sacrificial ceremonies on the family hearth 

a. A milk-tooth is thrown into the fire, whilst the child 
says »Locke, ge mig en bentand for en guldtand» 
or a corresponding formula. This custom is only 
found in Sweden 2 and it has spread even to Swedish 
Finland 3 and been adopted also by the Finnish 
speaking inhabitants. 4 

b. The membrane of boiled milk is thrown into the 

1 Loke i nyere overlevering in Danske Studier 1908, p. 193—207 
and 1909, p. 69—84. 

2 Hylten-Cavallius , Warend och Wyrdarne 2 I p. 176; Olrik , DS 
1909 p. 78—79. 

3 G. Landtman , Folktro och Trolldom (Finlands Svenska Folk- 
diktning VII, 1) p. 726—729. 

4 N. Setala , Finnisch-Ugrische Forschungen XII, p. 251. 


Jan de Vries, The problem of Loki 

FFC 110 

fire; the custom is known in Norway 1 as well as in 
Sweden. 2 

c. Some flour is thrown into the fire, when porridge 
is made. Only known in Telemark. 3 There are how¬ 
ever more sacrifices of this kind. 4 

d. The idea, that a spirit, called Loke, dwells in the 
fire is moreover suggested by the formula, said 
when the fire makes a crackling noise: »Lokji giver 
sine born hug» or »Lokje dengjer bon’e sine». Only 
known in Norway. 5 

II. Natural phenomena 

a. When in spring or in the beginning of summer the 
air is seen vibrating by the heat of the sun, the 
Juttish peasant says: »Lokke driver idag med sine 
geder» (West-Jutland, Skaane and Bornholm) or 
»Lokemand saar sin havresred» (Northern- and 
Middle Jutland). 6 

b. The sulfurous smell after a flash of lightning is 
called on the Faroes »lokadaun>>.' ! 

c. In the same islands Sirius is called lokabrenna. 8 

d. In Iceland the saw is known »Leingi geingr Loki 
ok ])6r, lettir ei hri5um». Probably this denotes 
that the bad weather will not come to an end. 9 

1 Storaker, Tiden i den Norske Folketro, Norsk Folkeminnelag 
II, § 462. Sometimes the words were said: »De’ ska Lokje ha’!» 

2 Bachtold-Staubli, Handworterbuch des deutschen Aberglau- 
bens III, c. 1762. 

3 R. Berge, M. B. Landstad (Norske Folkeminnesamlarar II, 2, 
Riser 1920) p. 18. The girl said, when asked why she did so: »login 
laut au hava litt». Perhaps login stands for Loki. 

4 Cf. Olrik-Ellelcilde, Nordens Gudeverden p. 255 sqq. 

5 Olrik, DS 1909, p. 78; Storaker , Elementerne i den norske folke¬ 
tro (Norsk Folkeminnelag X) p. 79. 

6 Olrik, DS 1909, p. 71—72. 

7 Olrik, DS 1908, p. 204. 

8 Olrik, ibid. 

9 Olrik, DS 1908, p. 206. 

FFC 110 

Chapter XI, Loki and modern folk-lore 


III. Loki in connection with weeds 

In Denmark different kinds of weeds are called Lokes 
gross or Lokes havre. 1 When it is nice weather in spring 
the peasants in Hedemark say: »Loke saar havre». 2 For 
Jutland see the examples under II a. 

IV. Loki and the spider 

In Sweden this insect is known under the names of 
locke, lock and the cobweh as locka-nat , lockasnara . 3 

V. Loki and the vermin 

In Telemark it is believed that Lokje brings lice on a 
sledge on Maundy Thursday. 4 

VI. Loki as a deceiver 

The only instances are two phrases quoted by Peder 
Syv: at fore Lokkes hr eve and at here paa Lockens even¬ 
ly r. In Iceland a big lie is called a lokalygi. 5 

2. 01 r i k’s hypothesis about 
the fir e-d e m o n Loki 

In his above-mentioned study Olrik has paid particular 
attention to the examples quoted by me under I a and b, 
II a. The sacrifice to the fire on the hearth, combined with 
the name of Loki proves that according to Norwegian and 
Swedish conceptions Loki resides in this fire and that he 
is a representation of it. Danish traditions on the contrary 
show him as a lysflimre-vcetten, which may be rendered in 
English as a spirit connected with the phenomenon of the 
summer-colts. He tries to combine these widely different 

1 Feilberg , Ordbog s.v. 

2 Visted , Vor gamle bondekultur p. 328. 

3 H. Celander , Lokes mystiska ursprung p. 18—26. 

4 Visted l.c. Skjaertorsdagaften skulde man vri tre pisker til 
at saette istand slseden for Lokje, som da kom kjorendes med et laes 
lopper og hadde kjort slaeden itu, da laesset var meget tungt. 

5 DS 1909, p. 78. 


Jan de Vries, The problem of Loki 

FFC llo 

conceptions by supposing, that there has been from very 
remote times the conception of Loki as a small spirit, whose 
name means »flame» or »fire-spirit» ( lysmand ). In Denmark 
he was thought of as luftflimmeret, in Norway and Sweden 
as the fire on the hearth. Out of these conceptions he gets 
the character of a teasing and thievish being. Here we must 
seek for the connection of Loki, for he is also of small stature 
and renowned as a trickster and a thief; hence — and here 
Olrik formulates his conclusion in the most prudent way — 
»Traits of the spirit of popular belief must have passed over 
onto the god Loki and contributed to give him a queer and 
complicated personality^ 1 

Against these conclusions serious objections may be 
raised. In the first place the connection between the god 
Loki and the fire-demon is not at all evident. The likeness 
is exclusively the teasing and thievish characteristic of 
both. Now this characteristic is not at all predominant in 
the popular traditions and Olrik can not do more than 
assert that it grew out of the conception of the fire-spirit. 
But then we may say that such a characteristic is not a 
special quality of a fire-spirit, but of the spirits in general; 
so the teasing Loki belongs to the wide group of the vcetter 
and shows no particular affinities with the spirit on the 

But even the idea of an original fire-spirit seems to me 
highly questionable. I must confess that I can not make 
any clear representation of such a spirit, when it could 
develop either into the hearth-spirit or into the pheno¬ 
menon of the summer-colts. For a fire-spirit is not a demon 
in one way or another connected with the abstract idea of 
fire, light, vibrating air or what else we may wish to imagine; 
it is supposed to be present in a definite form of fire. More¬ 
over a mythical view of a natural phenomenon is quite a 

1 DS 1909, p. 84. 

FFG no 

Chapter XI, Loki and modern folk-loiv 

22 0 

different thing from a spirit to which sacrifices are made. 
In fact, the ideas underlying both conceptions may be 
originally quite independant from the idea of fire. 

It will be worth while to discuss these forms of popular 
belief and to enter into details. Children are said to throw 
a milk-tooth into the fire with the words that Lokje may 
take it, but that he must give in return a tooth of gold. 
The custom and the rhyme are equally well known in all 
parts of the Teutonic world. But it is only in Sweden that 
the name of Lokje enters into the formula. In Norway they 
say e.g.: 

Ell, ell, no ska du faa i beintonn 

G-je so meg att i gulltonn. 1 

From Swedish Finland we have many examples of formulas 
with the names Lock or nock (both also names for the spider); 
besides we find mentioned gobb or onsgobb , that is »the old 
man of the oven». Or the formula is said in the form »Ge 
mej en bentann i stallet for en gulltann» without any part¬ 
icular reference to a spirit. 2 

The ordinary form of the formula however does not 
contain the name of Loke or a kindred being, but of a mouse. 
This is found everywhere among the Teutonic peoples. In 
Denmark it is not less common than in the other Scandi¬ 
navian countries 3 ; in Germany and Holland it is also the 
usual form. 4 Furthermore examples are collected among 
the Slavonic, the Finnish, the Lithuanian and the Lettish 

1 P. Lunde, Kynnehuset (Norsk Folkeminnelag VI, p. 33). Cf. 
also H. O. Opedal, Makter of Menneske, ibid. No. 23 p. 119. The 
first tradition is from Vestagder, the latter from Hardanger. 

2 G. Landtman, Folkloristiska och Etnografiska Studier III, 
p. 17—18. 

3 Examples are collected by Feilberg, Ordbog II, p. 631. 

4 Cf. Wuttke, Der deutsche Volksaberglaube der Gegenwart 
p. 351. 

230 Jan de Vries, The problem of Loki FFC 110 

peoples. 1 2 Therefore we must conclude that the formula, 
with the name of Loki is only found in a very restricted 
part of the wide area where the rhyme itself is current; 
the Swedish formula is an exceptional and not a traditional 
form. We must assent to the opinion expressed by the 
Danish scholar F. Ohrt a , that Loki has been introduced 
into a formula, that itself is surely not of genuine Scandi¬ 
navian origin. 

In fact, the sacrifice has nothing to do with the element 
of fire nor properly speaking with a fire-spirit. I have 
treated this subject in the French periodical Revue anthropo- 
logique. 3 Here I have drawn attention to the fact that in 
a great deal of examples it is stated that the tooth must 
be thrown behind the back or over the shoulders. Now 
this way of throwing an object backwards denotes clearly 
the character of the sacrifice: it is given to the spirits of 
the dead. The connection with the hearth-fire is then 
easily explained; here the family-ancestor, hustomten, is 
believed to reside and the sacrifice is not intended for the 
fire itself, nor for the fire-spirit, but for the tutelary-ghost 
of the family. In the same way the mouse may be explained 
as .a manifestation of the spirits of the dead, and that among 
the many possible forms (besides the mouse also the serpent, 
the butterfly, the bird) just this one is chosen, is obviously 
due to the fact that this rodent has such enviable sharp 
and sound teeth. 

So the sacrifice of the tooth has originally nothing to 
do with the heathen deity Loki. We might ask if the throw- 

1 Cf. Slansikha, FFComm. No. 87 p. 9 note. 

2 Trylleord fremmede og danske, Danmarks Folkeminder No. 25, 
p. 99. —. Likewise H. Celander l.c. 53 says that this tradition 
about Loki is only a petrified formula without any living meaning 
and that it is therefore most prudent not to build any hypothesis 
about the mythical nature of Loki on this custom. 

3 Le Jet de la Dent, Revue anthropologique XL (1930) p. 87—89. 

FFC 1 lo 

Chapter XI, Loki and modern folk-lore 


ing of the membrane of the milk or of some flour is to be 
interpreted as a sacrifice to this god. First we must dismiss 
the latter example, for the tradition quoted by Landstad 
only speaks of fire and not at all about Loki. The meaning 
of the former custom is not clear to me, as I do not know 
if the membrane is considered as a desirable or as a useless 
part of the milk. The latter supposition seems to me the 
most probable and in this case it is to be doubted, if we have 
the right to speak of a sacrifice; it may simply be a means 
of annihilating it. 

If we look at corresponding practices current among 
kindred peoples, we arrive at the same conclusion. In 
Germany the bristling noise of the fire is said to mean 
quarrel or calumny; if there are sparks on the hearth this 
indicates a visit. Such innocent omens are, of course, no 
relics of heathen belief. Then there exists the custom of 
feeding the fire, by throwing some morsels of every meal 
(in Bohemia) or some crumbs of bread into it. Sometimes, 
however, it is expressly stated, that this sacrifice is intended 
for the poor souls in purgatory, a christianised form of the 
idea, that the dead in general, but originally the dead 
members of the family, residing in the ancestral hearth, are 
honoured by such a sacrifice. Moreover customs, like those 
that the new-married woman is led round the hearth or 
that the fire is extinguished when there is a death in the 
house, clearly presupposes a very close connection between 
the fire on the hearth and the spirits of the ancestors. 1 
Folkloristic evidence of this kind makes it fairly certain, 
that in Scandinavia the same belief must have been pre¬ 

In any case we learn from it that in Scandinavia the 
name of Loki was combined with fire. But this connection 

1 Cf. Wuttke, Der deutsche Volksaberglaube der Gegenwart 
§§ 294, 430, 458, 566 and 737. 

Jan de Vries, The problem of Loki 

FFG 110 


is only limited to some regions and seems to be crystalized 
in a couple of traditional phrases. It would be arbitrary 
to conclude from it, that Loki was mentioned because 
originally he has been a fire-spirit. 

Still more difficult is the Danish belief about Lokke as 
a name for the summer-colts. It has been observed by 
Celander \ that Lokke is not the name for the undulating 
Warm air, but that this is called his cattle ( hjord , geder, 
faar ). We are reminded of the Frisian expression do summer- 
kalte lope as well as of the English summercolt. In Bornholm 
people speak about Kullebondens svin. Therefore we may 
surmise that the glittering, undulating air seen above a 
cornfield has made the same impression upon the people 
as the wind running through the cornstalks and making 
billowing furrows in them. They found for it the simile 
of animals moving through or above the field. Cattle, how¬ 
ever, suggests a being, who ownes and takes care of them. 
This being may be called kullebonden or bjergmanden, even 
Lokke, Lokkemand, Lukas or Jakob Loj 1 2 ; it is obvious that 
he is only a vcette supposed to live on the spot, where the 
phenomenon is observed. The name Lokke, then, does not 
belong to it from the very beginning, and it is still an open 
question what the name originally may have meant. 

Sometimes the phrase is Loke sows oats; we found this 
form in Jutland as well as in Hedemark. The warm air, 
vibrating in spring above the field, denotes that it is ex¬ 
cellent Weather for the growing of plants, to be sure not 
only of useful ones but also of Weeds. We have already 
seen that the Juttish people speak about Lokes grces and 
Lokes havre. But when We see that the Swedish names are 
trollhavre, pukhavre or liotfagar, we will not explain this 
with Lundgren by a substitution of the devil for the Wicked 

1 L.c. p. 55. 

2 Cf. Feilberg, Ordbog II, p. 445. 

FFC 110 Chapter XI, Loki and modern folk-lore 233 

Loki 1 , but on the contrary by assuming, that Loki is a 
particular name lor the bad spirit, to whom the origin of 
all nuisible plants and animals is ascribed. 

The case is very clear when Loki is said to bring vermin. 
For here he takes the role of the Christian Beelzebub, the 
god of flies and it is only the reminiscence of the Loki- 
figure in the last form of its development, that has led to 
this identification. It is not even necessary to accept any 
connection with those Old-Norse myths where he takes the 
shape of a fly or. a flea. 

Our conclusion is, that Olrik has built his hypothesis 
upon the frail base of popular traditions, where the name 
of Loki has been introduced afterwards. This explains 
why he is brought into connection with such very different 
phenomena as the sulfurous smell after a flash of lightning, 
the fire on the hearth, vermin, weeds and vibrating air on 
a hot day. If the idea of a fire-demon had been the root, 
from which modern popular traditions about Loki sprung, 
they would not have been so utterly disconnected. And 
when we wish to explain, why he has been introduced into 
these connections, we must not seek the solution of the 
problem in his fire-nature, but in some other element, that 
made him liable to such avatars in a time, when Loki was 
no longer a heathen god but only a vague recollection, 
hovering in the minds of Christianised people. 

3. The hypothesis of Hilding Celander 

In my criticism of Olrik I have trodden many a time 
in the footprints of Hilding Celander , who has given a 
vigorous criticism of the theory about the fire-spirit. I 
could leave out of the discussion several minor points, that 
are sufficiently treated in his paper. He urges that the 

1 Sprakliga intyg om hednisk gudatro i Sverige p. 80. 


Jan de Vries, The problem of Loki 

FFC 110 

Lokje of popular tradition bears all the characteristic traits 
of an elfish spirit, a vatte, nisse, bjergmand, or whatever you 
may like to call them. Hence he is of opinion, that Loki 
originally has been such a spirit and that not only popular 
traditions but even a few hints in mythological literature 
give sufficient evidence for it. Of course they do. But it 
does not prove the thesis of Celander. For if we choose in 
stead of well defined mythical beings of a lower order a 
kind of greatest common measure, in which they may be 
comprised, it is evident that a sharp formulation of the 
problem is altogether impossible. To explain the Loki of 
two thousand years ago Celander places himself on the 
standpoint of the modern representations about the vatte. 
But here we find only a vague and confused amalgamation 
of originally clear-cut mythological beings each with its 
own activities in its own domain. I may refer the reader 
to the study of G. Landtman \ who frequently stresses the 
fact, that the names for a certain being are numerous and 
variable. The story, that a person is bidden by a mythical 
being to assist at the child-birth of his wife, is told without 
distinction of »underbyggare, tomtar, bergtroll, vatten- 
varelser, r&d, troll och obestamde vasen». I should like to 
propose as a name for these »vague beings» the word vatte, 
which has become in course of time, and surely in folk- 
loristic treatises, devoid of all concrete meaning. 

By this conception Celander considerably weakens his 
attack on the hypothesis of Olrik. He is, of course, quite 
right, when he says that the Lokje of Telemark is a broader 
conception than »hearth-fire», but more in general a vatte, 
but it does not follow that Loki, who may be represented 
by this Norwegian Lokje, should have been likewise such 
a vatte. The vague ideas about this being may be as well 

1 Hustomtens forvantskap och harstamning, in Folkloristiska 
och etnografiska Studier III (Helsingfors 1922). 

FFC 110 

Chapter XI, Loki and modern folk-lore 


the result of the disintegration of popular belief in relatively 
modern times and it is quite possible, that a thousand years 
ago this Telemarkian Lokje has been a spirit of the hearth- 
fire in the strictest sense of the word. 

Starting from this basis it is easy for Celander to find 
proofs for his explanation everywhere. Loki is a typical 
elfish being in as far as he is a master in transformations, 
sudden disappearances and tricks of all kind. Scandinavian 
mythology even knows him as a small and nimble being, 
for the Soria I attr describes him as ekki mikill voxtum. And 
Celander argues in the following way: although the t>attr 
is not very reliable, this particular detail must be con¬ 
sidered as original. Why? Because it fits well into his 
theory? But then he leaves out of account and does not 
think it worth while discussing, that there are more possible 
explanations for the small stature of Loki in this fabulous 
story of the 12th century. I have already mentioned the 
fact that according to Olrik’s epic law of opposition the 
strong and huge Thor must be accompanied by a nimble 
and witty being. J>jalfi who is dragged through the stream 
hanging at Thors belt and who in another myth finds an 
excellent stratagem in the battle against Hrungnir, shows 
the same qualities as Loki, when he accompanies Thor to 
Thrymr or when he is dragged through the air by an eagle. 
The small size of Loki is then only a motive of epic tales. 
But there is also another reason to distrust the story of the 
Sorlaj attr; in the time when this was composed heathen 
religion had disappeared and Christianity superseded it. 
This may have caused a change in the conception of Loki 
and then of course this has taken place in a line of develop¬ 
ment, of which the Telemarkian Lokje and his other Scandi¬ 
navian colleagues form the end. Is there any reason to 
take this descriptive sentence of the pattr as valuable 
evidence for the original conception of Loki? Putting the 
question is at the same time giving its answer. 


Jan de Vries, The problem of Loki 

FFC 110 

For Celander the most interesting point of the cycle of 
popular traditions is the Swedish name for the spider: loche 
and for the cobweb: lochanat or lockasnara. This gains part¬ 
icular importance because in the Snorra Edda a myth is 
related about Loki inventing the fishing-net. And because 
the cobweb is also called dvarganat, we may conclude that 
the name goes back to a conception of Loki, if we presuppose 
at the same time, that the locke of popular belief has been 
a kind of vatie, related to the dwarfs. If, however, any 
nearer relation must be assumed between the spider and the 
god Loki, this should not depend only on the similarity of 
the name, but also on the mythical or religious representa¬ 
tions attached to it. We might have expected that Celander 
would try to prove that either Loki originally has had some 
particular connection with the spider, or that cobwebs have 
been considered as an invention of Loki. Celander obviously 
shrinks back from this conclusion and in the course of his 
study he undoes the knot he has tied so vigorously at the 

The cobweb is called locJcanat because a supernatural 
being called Locke is supposed to have invented it. The 
reason is because the dwarfs and other elfish beings are 
clever artisans and because on the other hand the miracle 
of the wonderful cobweb, in autumn being hung my¬ 
steriously during the night between the trees and the shrubs 
demands a special explanation. The popular fantasy is 
answered by the myth, that the dwarfs or the elves have 
done it. Quite possible and it may even be granted that 
such a mythical interpretation is of exceedingly remote 
antiquity. But why must the connection of Loki and the 
cobweb have been of the same hoary antiquity? Now we 
might expect Celander to adduce the argument of the name. 
The spider is called Locke and its web lockanat; hence these 
words presuppose the identity of Loki and the word locke 
for spider. Here again Celander does not venture to draw 

FFC ill) 

Chapter XI, Loki and modern folk-lore 


this conclusion. Although he -derives both words from the 
verb lukan, he arrives at two totally different conceptions. 
The word Loki means »the closer», but the being that com¬ 
prises something is the spirit of the mountain (bergmannen), 
where the dead are gathered, or as we may say in the well- 
known terminology: a vatte. And the spider is called locke 
because it encloses, ensnares insects. 

What then remains of the alleged identity of the spider 
and Loki? If there is on the one hand a vatte with the name 
locke and on the other hand the spidername locke, there 
is no reason why the identification should not have taken 
place at every period in heathen as well as in Christian 
times. It would have been an indication of no slight im¬ 
portance, if Celander had been able to prove, that a word 
as lockanat means the Web spun by a vatte, that was called 
locke. Then the word locke for spider would have been 
derived from the composition lockanat , where people did 
not feel any more the right meaning of the first component. 
But Celander only adduces reasons of linguistic as well as 
of psychologic character for the subsequent identification 
of the two words. Why then should it be necessary to accept 
in the spider-name locke the same word-root as in Loki? 

For the name of the heathen god and its modern con¬ 
tinuations we observe forms with k and fefc; the same is the 
case for the spider-name. Celander pretends that there is a 
complete identity between both words as to the geographi¬ 
cal distribution of the forms with k and kk\ in the Eastern- 
Scandinavian countries we find locke , in the Western area 
however loke. It is only possible to make such an affirmation, 
when we consider the isolated form Loye Langbein, used in 
Telemark for the spider, as derived from a word loki 1 . The 
hypothesis that in a position of feeble accentuation the sound 
k may pass into j in some Norwegian dialects, is indeed 

1 Cf. Ross, Norsk Ordbog p. 486. 


Jan de Vries, The problem of Loki 

FFC 110 

scarcely sufficient to prove 1. that in this case the same lin¬ 
guistic development has taken place, 2. that in this name 
the word Loye means »spider» 1 and 3. that this name has 
been known in the whole extent of Western-Scandinavian 
language. As long as this is not made plausible, we may 
regard the isolated name Loye Langbein as very suspect 
and we are obliged to arrive at a slightly different conclus¬ 
ion from that of Celander, viz. 1 . The name locke for spider 
is only known in Sweden, 2. In Sweden and Danmark an 
elfish spirit is called by the same name. If this locTce is the 
same as the Old-Norse Loki then the conclusion lies near at 
hand, that the fcfc-form in Sweden is due to a secundary 

The etymology of locke »spider» proposed by Celander is 
of course only one out of many more possibilities. Loki 
may represent a Huken, nom. agentis of a verb lukan »to 
closes. But the fundamental meaning is »to bind, to twists, 
as seems to be proved by on. lokkr, oe. locc, as compared 
with gr. luginos sfletteds and lit. lugnas sflexibles. Then a 
word locke may be derived from a w-stem *lugen gen. Hugnos 
giving the corresponding forms loki and lokki. But this 
word may have meant sthe twister, the benders, not sthe 
encloser, the ensnarers. Hence if we should accept the 
etymology of Loki as sthe enclosers (and it must be emphasi¬ 
zed that this is only a hypothetical explanation of the name), 
the relation with the spider’s name is almost nihil, for if the 
word locke goes ultimately back to the same root lukan , 
the meaning may not have been the same in both cases. 

It is exceedingly dangerous to build a theory about a 
religious phenomenon upon the quicksands of etymologies. 
Because Celander himself is not able to make use of the ety¬ 
mology of locke for his hypothesis about Loki’s character 

1 Loye Langbein is compared by Celander p. 24 himself with a 
Swedish Halge Hogben. Then Loye can be something quite differ¬ 
ent from a word for spider. It may be f.i. a proper name. 

FFC 110 

Chapter XI, Loki and modern folk-lore 


as an original vatte. we may dismiss it from our discussion. 
But then the last support of his theory has given way. If 
the similarity between the names of Loki and the spider 
is not original at all, the likeness between the cobweb and the 
fishing-net invented by Loki dwindles into a mere accident. 
The conception of a vatte, in the modern sense of the word, 
is far too vague and broad to give us the clue to the real 
meaning of Loki. 1 The results, arrived at by Celander, are 
not more convincing than those of Olrik; methodologically 
his investigation is undoubtedly inferior to that of the Danish 
scholar. It is always better to start from a well circumscri¬ 
bed and tangible religious conception as e.g. a fire-spirit, 
than from such a wild and incoherent medley as the modern 
representations about the »vatte». 

4. The value of the folkloristic material 

If I would not trespass beyond the boundary of this 
study, which has expanded already into considerable length 
on account of the polemical digressions I was obliged to 
start upon, I might have done more justice to the studies of 
Olrik and Celander than the case has been in the few pages 
I have been able to spend upon them. Nevertheless I hope 
to have succeeded in showing that the explanations proposed 
by both scholars are quite unsatisfactory. It would not be 
expected otherwise, because they laid their superstruction 
on a bad foundation: popular belief of modern people is a 
very inappropriate material to make use of, if we want to 
explain religious conceptions of heathen times going back 
about a millenium. I came to the same conclusion in my 

1 Celander seems sometimes to be inclined to make a more con¬ 
cise definition of the vatte, as a »spirit of the dead» (f.i. on p. 71), 
but he does not clearly express himself as to the question, if Loki 
has been a ghost, or an elfish spirit in general. 


FFG 110 

Jan de Vries, The problem of Loki 

study on the popular traditions about Othin 1 and since I had 
the opportunity of treating it more theoretically elsewhere 2 , 
I may think myself relieved from the necessity of repeating 
the same arguments. 

But one may expect me to explain the obvious reminis¬ 
cences about Loki in modern popular lore. Of course I shall 
try to make it probable that these are no original and undis¬ 
torted specimens of a very primitive conception of Loki, 
even older and more genuine than the sources dating from 
heathen times; on the contrary they are to me worthless 
deformations, that can not give us any idea about the con¬ 
ceptions that underlie the Old-Norse deity Lola. 

First I wish to consider the very disappointing results to 
which Olrik has come on the basis of his investigation. 3 
There are four different Loki conceptions, which I already 
had the opportunity of mentioning: 1. The firespirit of 
popular tradition, 2. The companion of Othin, an original 
culture-hero, 3. The companion of Thor, the servant of the 
thundergod, 4. The devilish Loki. If we ask in what way 
the first meaning, based on the traditions of modern popular 
lore, can have had any influence on the Loki-figure of heathen 
mythology or at any rate may be discovered in some of its 
particularities, we get the following answer: the Lokke- 
figures of popular belief, that is the two very distinct forms 
of the fire-spirit in Norway and the »Lysflirore-v8etten» of 
Danish tradition, are both exceedingly old 4 : »They may, in 
some respects, have exerted an influence upon the small 

1 FFComm. No. 94. 

2 Die Bedeutung der Volkskunde fur mythologische und religions- 
geschichlliche Untersuchungen, Germanisch Romanische Monats- 
schrift XX, p. 27—39 and Volkskunde en Volkenkunde als hulp- 
wetenschappen der Godsdienstgeschiedenis, Mensch en Maatschappij. 

3 Festskrift Feilberg p. 587 sqq. 

4 Vi har altsaa to naturfodte Lokke-typer af a-ldgammelt ud- 
spring: mytiske opfattelser af tilvserelsen i den naiveste form (p. 586). 

FFC 110 

Chapter XI, Loki and modern folk-lore 


size and, teasing character of Loki, and perhaps also on his 
thievish and even wicked character, although this last 
quality is not very important in the popular belief». In this 
sentence, a literal translation from the original, lies the 
total result of Olrik’s study about the Lokke of folk-lore. 
In some respects it may have had influence upon such insigni¬ 
ficant details as his small size or his teasing character, that 
evidently demands a more satisfactory explanation in other 
respects as well. But the role of Loki as the servant of 
the thundergod, was it not quite enough to explain his small¬ 
ness and did the Esthonian tale, adduced by Olrik, not prove 
that he was a clever-minded person? I get the impression, 
that Olrik, who saw no means of introducing the fire-spirit 
through the main door, has tried to slip him in through the 
back-door. Had we not better leave him standing altogether 
outside, if he is only able to exert a very restricted influence 
»in some respects»? 

But Olrik adduces a new argument for his quadruplica- 
tion. The four different constituve elements has each his 
special name. The fire-spirit is called Lokki, a side form to 
Logi; Lodurr is the name of the culture-hero; Loptr is the 
name of Thor’s companion; Loki finally is the fettered demon. 
I will not insist upon the arbitrary character of this distinc¬ 
tion, in no way borne out by the texts, nor upon the fact 
that there is no reason to identify L65urr with Loki, but I 
shall only follow in the footprints of Olrik. According to 
him Loki and Lokki may be hypocorystic names formed by 
the suffix - Id , the former from a name ending in one con¬ 
sonant, the latter from a word with two consonants. Hence 
Lokki as »Kosename» becomes quite accidentally the same as 
the name of the fire-spirit. In fact Olrik proceeds with the 
following statement x : »In this way we may find an explana¬ 
tion for the mythical coherence by means of the circumstance 

1 L.c. p. 590. 



Jan de Vries, The problem of Loki 

FFC 110 

that a divine thief of the fire (Lddurr) in course of time coales¬ 
ced with the thundergod’s servant (Loptr) and partly also 
with the early lightspirit of popular belief (Lokki)». What 
then is the use of the long and learned digressions about 
these religious conceptions, if we are obliged finally to con¬ 
cede that mere chance, a secondary similarity between divine 
names, has brought about the curious and complicated figure 
of the Eddie Loki? To believe in such an accident is to give 
up all hope of a real explanation. Should there be no religious 
elements that belong to the same sphere of conceptions, 
should there be no original meaning from which the diffe¬ 
rent characteristics of Loki could emanate, then a single name 
could not have worked the miracle of Loki’s personality. 
Even if we are inclined to accept with Andre Jolles that Loki 
has become a kind of basin into which mythological tales 
of all kind have been collected, then a name alone is scarcely 
sufficient to serve as a rallying-point; there must be also in 
the conception itself an element, that could attract such a 
variegated mass of tales, as is the case with Loki. This mag- 
netical central point is, in my eyes, the teasing character of 
the god; the trickster type is liable to expansion in diffe¬ 
rent directions and funny tales about deceivers and robbers 
may at all times be transferred to it. But again I must insist 
upon the fact, that such a trickster-type should not be con¬ 
fused with the elfish demons of popular belief. 

How then must we explain the different beings of popu¬ 
lar lore that are called by names reminding one of Loki? 
If we consider the modern representations as the result of 
a secondary development, there is no reason to suppose 
that the different forms, as they are found in Norway, 
Sweden, Denmark and elsewhere, should be explained in 
the same way. It is even more probable, that under widely 
different circumstances the results of combinations and 
alterations must have been very divergent from each other. 
We may assume, e.g. that in countries as Norway and Ice- 

FFC110 Chapter XI, Loki and modem folk-lore 243 

land, where literary sources show a richly variegated tra¬ 
dition about Loki, its impress upon now-a-days popular tra¬ 
dition must he greater than in Denmark and Sweden, where 
the myths about Loki may be supposed but can not be 
proved to have existed. 

The sacrifice of the tooth may be adduced as a first 
example. Why has Loki in Sweden been introduced into the 
formula of the milk-tooth? Obviously because people felt 
some connection between him and fire. Fire, as is well 
known, is always conceived of as a living being; the play 
of the flames, the noise of the burning wood, the dreadful 
danger of a conflagration are reasons enough for such a 
natural conception. Sometimes they speak in Norway of 
varmeltjeninga or huldra; a very meagre form of personifica¬ 
tion. A more colourfull expression is Eldbjer denger barna 
sine, a parallel to the already quoted phrase Lolcje dengjer 
sine drengjer. Eldbjer or in an older form lldbjerg is a very 
amusing saint; for she is only the result of a queer misunder¬ 
standing. At the end of Yuletide there were formerly special 
ceremonies on eldbjorgs-day, the day when a sacrifice was 
made to the fire in order to prevent it from breaking out 
into conflagration during the next year. The ceremony 
intended to cover up the fire was called ildbjerg , and people 
were wont to drink a special mug of beer or as they called it 
an Eldbjersskaal or -minde. This was the reason why Eldbjer 
was in later times considered as a particular saint to be 
venerated on this day and particularly connected with fire. 1 
It is this fictitious saint, who gives a thrashing to her chil¬ 
dren, when the fire makes particular crackling noises. 2 

1 Gf. Yisted, Vor gamle bondekultur p. 242 sqq; Olrik-Ellekilde, 
Nordens Gudeverden p. 256 sqq. 

2 This is, of course, only a tale to explain the crackling noise, 
that is apt to attract people’s attention and curiosity. When in Lesbos 
the wood begins to whistle in the fire, people say »Be silent»; if it 
obeys, an enemy is supposed to have abused you, if it does not, a 


Jan de Vries, The problem of Loki 

FFC 110 

It may be that it is some misinterpretation of an elder eldmor, 
that is found at the side of it, or it may be that from the wish 
to personify the spirit on the hearth, people chose Eldbjer, 
because her name contained an element meaning fire. But 
then, why should they not identify Loki and the fire logi, 
which, as we have seen, induced even a mythographer of the 
Middle Ages to make a pun in the myth of G-eirrndr? And 
if so, what is then the real relation between Loki and fire, 
unless a superfluous coincidence of names? 

So I reject the name of Loki in the tooth-formula and 
other kindred sayings as quite untraditional. But granted 
that an original connection should be admitted here, are we 
then anywhere nearer the problem, what the real meaning 
of this deity has been? Surely not, for in the practices, nowa¬ 
days connected with fire, different conceptions and beliefs 
are inextricably commingled. Let us take an example: 
at present in case of illness often a sacrifice is made to the 
fire. We have mentioned the milktooth. Also part of the 
clothes or of the medicine is thrown into the fire. This is a 
sacrifice to the fire or to the spirit of the hearth ( drevetlen ) 
as Reichborn-Kjennerud says. 1 But what kind of spirit? 
Indeed a being representing fire? Or the spirit of the hearth 
or of the house, residing on the fire-place? Or the spirit of the 
ancestor, supposed to dwell in the same place? It is a well- 
known fact, that the spirits of the dead may cause illness of 
all kind and that they consequently have the power to cure it 
also. 2 The sacrifice to the fire in case of illness has nothing 
to do with this element, but is intended for the ancestor- 

friend has spoken good of you (G. Georgeakis et L. Pineau, Le Folk- 
Lore de Lesbos p. 352). Perhaps the practice mentioned in the 17th 
titulus of the Indiculus Superstitionum et paganiarum viz. de obser¬ 
vation pagana in foco, has been of the same kind. 

1 V&r gamle trolldomsmedisin p. 150. 

2 Cf. my Van Alven en Elfen in Volkskunde 1931, p. 14 


FFC 110 Chapter XI, Loki and modern folk-lore 

spirit, who has taken its abode on the hearth. 1 But this is 
surely not the meaning in all the cases enumerated by the 
Norwegian scholar. The first bandage on a wound is thrown 
into the fire. Is it properly speaking a sacrifice? It is only a 
means to annihilate an object, that has been in magical 
contact with the wound; the first bandage was perhaps 
supposed to contain such bad matter as would cause inflam¬ 
mation (observe the similarity of notions between inflamma¬ 
tion and flame) and then it was most expedient to destroy 
it utterly. When a child has for the first time got epileptic 
spasms or inflammation of the lungs, part of its clothes are 
burnt. Again we ask: is this a sacrifice? It is evident that it 
is not: these clothes contain part of the ill person, as its nails 
or hair or shoes or whatever you like, and if they are burnt 
the illness itself is destroyed. 2 If you want to render a witch 
inoffensive, you take a waxen image and then you burn it 
slowly; surely the witch will succumb to intolerable suffer¬ 
ings. Why should not one take a part of the clothes infected 
by the disease, and throw it into the fire in order to annihilate 
the illness itself? Sacrifice is an easy word for all such prac¬ 
tices, but it would be advisable to use it only, when we are 
sure that the action is intended to influence a well-defined 
being, residing in the place where the sacrifice is made. 

Loki as a kind of fire-spirit has nothing in common with 
the Loke who sows weeds. In expressions as Lokes gross or 
Lokes havre he is of course only a representative of the devil. 
The same is the case when he is said to bring vermin. Here 

1 Sacrifices to dead ancestors are often made on the hearth, as 
e.g. those to the Stopan in Bulgaria (cf. Jan Machal, Slavic Mytho¬ 
logy in Mythology of all Races III, p. 238), who is supposed to be 
descended from an ancestor distinguished for valour and bravery. 

2 In Finland some salt is rubbed against a tumour and then 
thrown into the fire whereupon the person jumps back from the 
hearth so as not to hear the crackling noise of the burning salt. This 
act is accompanied by the words sy-ylat palaa or syylani palavat 
(cf. F. A. Hiistesko, Lansisuomalaiset tautien loitsut p. 15). 

246 Jan de Vries, The problem of Loki FFC 110 

we recognise Beelzebub in a new transformation. Here again 
there is no reason to look for old conceptions of Loki to 
explain these popular beliefs of later times. He is the bad 
Loki, who in the last phase of the Scandinavian mytho- 
graphers’ activity had been duly developed. We need not 
be amazed. Loki was a being of ambiguous reputation; when 
Christian creed declared that all the heathen gods were to 
be condemned as bad demons, Loki was certain to fall 
deepest of all. Moreover medieval monks, so clever in 
making the most absurd etymologies, surely have, noticed 
the similarity between Loki and Lucifer. If it is true that 
by a kind of primeval revelation primitive peoples may 
have got an idea of a. supreme god — as some ethnologists 
have pretended — then surely Teutonic religion had its 
devil avant le date. Loki, Well known from a series of very 
funny stories, could easily be kept in people’s memory and 
then he could make his appearance whenever it might seem 
possible to ascribe to him something that is dangerous or 
wicked. The close resemblance between the words Lokjen 
and Laakjen (the devil, from the adj. laak wicked, bad») 
has also possibly exercised an influence to turn the heathen 
name Loki into a word of ill omen. 1 Why was he brought 
into connection with weeds? Surely not because he had some 
kind of relation with agricultural pra ctices. But the explana¬ 
tion is as simple as possible: there are hundreds of tales 
about a creation of useful plants and animals by God and 
of nuisable ones by the devil. The famous collection of 
Dahnhardt gives a considerable amount of examples; they 
may be easily multiplied by running through the mythologi¬ 
cal tales of American-Indians, Australians or other primitive 
peoples. I only wish to quote one example 2 from Russia: 
when God and the devil together are creating the world, 

1 Cf. Visted, 1. c. p. 328. 

2 Dahnhardt. Natursagen I. p. 187. 

FFC110 Chapter XI, Loki and modern folk-lore 247 

God has sown the oats, the devil the weeds, God has created 
the cow, the devil the goat. Numerous nuisable plants 
have been called after the devil, from the pukhavre, already 
mentioned, to the duivelsnaaigaren and the Teufelsabbiss. 

Words like lokadaun and lokabrenna, current in Iceland, 
are likewise due to the circumstance, that on this island of 
old lore the recollections about the heathen gods were not 
easily forgotten. How else could a phrase have been trans¬ 
mitted to posterity like the one quoted: Leingi geingr Loki 
ok por, lettir ei hridum? 

In the theory of Olrik expressions as fore Lokkes breve 
did not fit in well. They are of course quite natural, if we 
consider Lokke as a devil, one of the numerous devils that 
invaded Scandinavia when the Christian religion conquered 
the North. 

But I must confess, that I am confronted with the diffi¬ 
culty of explaining why the summercolts are called the goats 
of Lokke. I should not like to affirm that Olrik has explained 
it from his standpoint in a satisfactory way; the vibrating air 
can scarcely be personified as a mythical being of »luftflim- 
meret». I have already shown that people explained the 
dancing hot air above the field in which the objects themsel¬ 
ves sometimes became invisible, and only moving and undu¬ 
lating waves of glittering hue were to be seen, by a flock 
of animals springing to and fro. These animals were of 
course of a supernatural kind; they belonged to a bjergmand 
or a kullebonden. But what is the reason that these creatures 
should have been so evilminded as to call them by a name, 
which could only suggest some evil trick of the devil? It 
seems to us so harmless, and even so pleasant, this floating 
of flashing waves on a warm day in spring. Obviously the 
peasant did not think so. Indeed the Juttish peasant says of 
the same phenomenon: Lokemand sar sin havresced. We know 
what that means: the devil is out to sow weeds in the field. 
We may compare with it a saw, current in Mecklenburg: 


Jan de Vries, The problem of Loki 

FFC 110 

Nu hoddt de Duwel sin Schap up Land, wenn dal so flammert, 
wenn de Luft so wackelt. Hence the peasant looks with some 
distrust at those glittering waves, for such a heat in spring 
gives the weeds too great an advantage and they may 
suffocate the grain, that is growing more slowly. This may 
he the reason, why Lokemand has been supposed to be in 
action during the warm hours of a cloudless day in spring; we, 
civilised people of the towns, only visiting the country for 
a short time of relaxation and pleasure, do not realise the 
dangers that are assailing the peasant, for whom nature is 
an earnest companion, but not less an unrelenting foe. 

The different popular traditions are useless for making a 
reconstruction of the original forms of the heathen belief. 
They are quite late and moreover debased and altered in 
their essential parts by so many ages of Christianity. But 
perhaps they may give us an indication as to the geographi¬ 
cal area where Loki was known in heathen times. We found 
recollections in all the Scandinavian countries and we may 
conclude from that, that he was known in Sweden as well as 
in Iceland, in Denmark as in Norway. He is not a special 
development of West-Scandinavian myth-making, although 
it must be conceded, that it is not likely, that the numerous 
tales told about him in Iceland ever have been known all 
over the Scandinavian world. We must be very carefull in 
our conclusions, and I do not venture to go farther than the 
assumption, that it is especially as a trickster, that he has 
played a r61e in the myths of the heathen Swedes and Danes. 

But this god was, as far as we. can see, not venerated 
by any cult or sacrifice; he was only the hero of curious 
tales, mostly tales of scandal. Why has he survived the 
deluge, that swept away so many precious things of heathen 
lore? Other gods may have been bound up so tightly with 
rites of fertility or of the commemoration of the dead that 
they could not be extirpated so easily. But Loki, whom the 
heathens themselves, arriving at a higher degree of civilisa- 

FFC 110 

Chapter XI, Loki and modern folk-lore 


tion, may even have looked upon with contempt and anno¬ 
yance? Certainly, Loki as a god could easily have been for¬ 
gotten. But the tales current about him were so very 
interesting and amusing, that they might please even an 
audience of Christians, who could feel themselves relieved 
by the idea that the new creed had freed them from such 
abominable superstitions. There is one example of such a 
tale preserved unto modern times; the popular ballads 
treating of Thor’s visit to the giant Thrymr. Here Loki 
acts as the cunning companion of the thundergod and the 
mediaeval ballads have not dropped his name; in Nor¬ 
way he is called Lokke Lagenson, in Denmark Lokke 
Leyemand, in Sweden Locke Lewe or Locke Loye 1 . Words 
like these, preserved in poems handed down from genera¬ 
tion to generation, are not likely to pass from the memory 
of a people; they are crystallizations of a series of old tradi¬ 
tions, mostly forgotten, but partly dimly recollected and 
they preserve moreover their typical character more preci¬ 
sely because a funny tale remains attached to it. Among 
people, who liked to hear about his adventures on his journey 
with the feigned bride of the giant, the memory of his 
name and of his deeds could not be lost altogether. 

There were, indeed, many motives for a preservation 
of Loki. In the last times of myth-making he had been a 
favorite of poets and story-tellers and a large cycle of tales 
had been collected around him. The new creed turned him 
into a devil, to which he had shown always a dangerous 
proclivity. Hence one of the devil’s names became surely 

1 The second element is not satisfactorily explained, cf. Bugge- 
Moe, Torsvisen i sin norske form p. 91sq, where the first opinion of 
Bugge (Lewe, Loye, Leimand etc. from Laufeyjarsonr) is given up, 
but a still more problematical explanation by means of the word 
Ldfturr is put instead of it. Is it not possible that the words are con¬ 
nected with lugi »lie» or dan. Idier. sw. loje (from on. hlegi) ’joke, 
fun, trick’? 


Jan de Vries, The problem of Loki 

FFC 110 

Loki; in fact it must have come to this, since the name remin¬ 
ded one so much of the terrible Lucifer. Besides the tales of 
old times were not forgotten; sometimes they were even re¬ 
newed so as to satisfy new exigencies of literary tradition. 
Finally after many vicissitudes and in the course of a gradual 
debasement he has come down to modern times: a shapeless 
figure and a kind of queer bogey; let us be merciful and call 
him a »Vatte». 


1. Tli •' present state of affairs 

In the preceding chapters several theories about the 
origin of Loki have been discussed. The available material, 
supplied by literary monuments and by folk-lore, has been 
treated, as far as possible referring to earlier publications 
the results of which were quite in agreement with our own 
views. We find ourselves at last in the presence of a bewilder¬ 
ing mass of contradictory theories, all however built up on 
the basis of data that are the same for every scholar. 

Loki is explained as a god, whose importance is rapidly 
dwindling away in the time, when the literary traditions 
begin to shed any light upon the peoples of Scandinavia. 
Others, however, say that he is a rather subordinate religious 
being, raising himself in course of time but never arriving at 
the dignity of a full-blown god. All scholars are at any rate 
of the same opinion with regard to the secondary character 
of Loki as the fettered demon and as a kind of heathen devil, 
although there is by no means unanimity about the way in 
which this development has taken place. But the rest 
that remains after cutting away the different accessory 
elements, is judged sometimes as an amalgama of various 
characters, that originally had nothing in common with 
each other. Whilst one scholar was intent upon assigning 
to the literary conception of Loki the largest possible amount 
of foreign, and particularly of Christian influence, others 


Jan de Vries, The problem of Loki 

FFG 110 

were willing to consider him as a deity, essentially heathen. 
The contradictory elements in his personality were explained 
as the result of a rather accidental collision between utterly 
independant deities, but on the other hand there were scho¬ 
lars who started from one original conception and explained 
the multivarious Loki of later times as the result of an histo¬ 
rical development. 

Loki as a god of fire, of fertility and of the underworld; 
Loki as a culture-hero and as a malevolent demon; Loki as a 
servant of the thundergod, as an elfish spirit or as a »luft- 
flimre-vsette»; the explanations are manifold and it seems 
simply a waste of time to try to explain a religious phenome¬ 
non, that has baffled the efforts of so many eminent scholars. 
Are we to add fresh proof of Loki’s inextinguishable ingenu¬ 
ity in deluding gods, giants and scholars alike? 

Still I am not quite so despondent about the possibility 
of understanding this deity. But it will be advisable to be 
careful not to fall into the errors, others have made before 
us. I exclude fundamentally the total amount of folkloristic 
material, and I consider as the only reliable source the liter¬ 
ary traditions of the early Middle Ages. Furthermore I am 
of opinion that we must not try to understand Loki as a 
kind of mosaic pattern, to which different ages and various 
peoples or even several religious conceptions have furnished 
the singular stones. Although it can not be denied, that 
there are Christian influences, which have distorted the real 
heathen Loki, we must try to find the central point from 
which he may be explained. 

This central point can only be found, if we are able to 
catch the real character of the god. Provisionally it is not 
necessary, it is even not advisable, to inquire into the origins 
of this religious phenomenon; we have only to ask: in what 
character did the heathen Scandinavians of the tenth 
century — to fix a date, when the conversion had not yet 
brought to an end the natural growth of pagan religion — 


FFC 110 Chapter XII, Loki as a religious phenomenon 

imagine Loki. The different humorous tales, heaped upon 
him by later mythographers, must then become intelligible, 
not as real trustworthy myths of course, but as the results 
of a development starting from this tenth century conception. 

It is admitted by all scholars that the most outstanding 
feature of Loki is his character as a trickster and a thief. 
With only very few exceptions all the traditions about 
him show him as a cunning creature, delighting in making 
mischief. Sometimes he shows a rather childish pleasure in 
playing his tricks upon the gods, often he contrives to do 
serious damage, but in most cases he is obliged to repair 
his faults. I have laid stress on the important fact, that he 
is not the declared foe of the gods; on the contrary he is 
closely related to them and in his relations to the giants 
there are no indications of his being their ally. He may give 
the impression of standing on the verge of the world of 
divinities, still we have ample evidence that he is rather 
gliding down from the bright realm of the Asir into the 
darkness of trolls and monsters, than that he is a newcomer 
among the gods, looked upon with reasonable distrust. 

Hence our task of finding the solution of the problem 
lies in the explication of this character of Loki. Scholars 
have been too prone to label him as a god belonging to the 
spheres of divine activity, a theoretical view on religious 
phenomena had set up as normalised for gods in general. 
Gods representing the forces of nature, the elements and the 
celestial bodies, gods of death and of fertility, gods that ans¬ 
wer to the needs of primitive peoples in the course of their 
daily life, such were the types outside of which no satis¬ 
factory explanation seemed to be possible. Even those, who 
clearly observed Loki’s character as a trickster, shrunk from 
making it the central point of the conception. Whenever 
an even trifling detail seemed to surpass this trickster-type, 
scholars eagerly followed the track it opened before them, 
hoping to detect a conception that either was more original, 


Jan de Vries, The problem of Loki 


or at least could better fit into the theoretical notions about 
primitive deities. 

Loki as a trickster is quite sufficient as a religious pheno¬ 
menon. Scholars as Olrik and Von der Leyen have paralleled 
him with the well known tricksters of the North-American 
Indians. Indeed, here is the possibility of fruitful compari¬ 
sons. When I give a short sketch of these American »deities» 
I do not intend to use them as a means of explaining Loki, 
for I am fully aware of the wide gulf that divides the aborigi¬ 
nal tribes of North-America from the heathen Scandina¬ 
vians. I only want to give a clear idea about the conception 
of the trickster-type itself. 

2. The trickster of primitive religions 

In nearly all tribes of North-America in some way or 
another a supernatural being is known, whom several curious 
stories are told about, that show his cleverness in playing 
tricks on animals and men, but also showing that he himself 
may be the victim of an others intrigues. If people some¬ 
times consider him with a certain respect, he is not less 
the hero of many a funny story, that is not particularly fit 
to enhance people’s opinion about him. Very rarely is there 
any apparent connection between the various tales and jokes; 
they form rather a wild mass of heterogeneous traditions. 

As an example we may take the Raven Txa’msem of the 
Tsimshian tribe of which P. Boas has given an accurate and 
full treatment. 1 Here we meet such tales as these: he makes a 
slave from a log of wood and makes use of him for his tricks. 
To procure fish he feigns death and is put into a coffin. But 
when they have got the fish they want, the slave eats it 
all himself and the poor remainder is left for the Raven. 
In this case the deceiver is deceived, but more often he has 

1 Tsimshian Mythology in the Report of the Bureau of American 
Ethnology XXXI (1909—1910) Washington 1916, pp. 60 sqq. 

FFC llo Chapter XII, Loki as a religious phenomenon 255 

more success, e.g. when he kills a hear by inducing it to cut 
pieces of flesh from its own body or when he pretends to 
build a canoe, but only cuts with his axe into a mouldered 
cedar and meanwhile consumes all the provisions. It seems to 
be a very good joke, when he tries to imitate the magic forces 
of other beings and by doing so injures himself, as when he 
cuts meat out of his legs, or dugs it out of his body by means 
of a sharp stick or an arrow, when he kills his children for 
food or when he expects oil to drip out of heated hands. 1 

These rather innocent merry tales, however, are not the 
only traditions about Txa’msem. His first deed was the 
robbing of daylight from heaven, where it was kept in a box 
and he succeeded by transforming himself into a cedar- 
leaf, that floated in the bucket from which the daughter of 
the chief of heaven was about to fetch water. She drank 
from the bucket and Raven was shortly afterwards reborn' 
as an inmate of heaven. According to another story he pro¬ 
cures fire with the aid of a deer; he causes the tides. But he 
is not always so friendly-minded, for it is through him that 
mankind has short life. 

These examples will be sufficient to show that the Raven 
in Tsimshian mythology is a very complicated figure. He is 
a trickster with a clear proclivity to triviality; as to his 
character he is greedy, selfish and treacherous. But evidently 
he is more than that: he is also a culture-hero who procures 
for mankind such indispensable things as sunlight and fire. 
He is scarcely an object of worship; he may rather be com¬ 
pared with those heroes of the past, about whom indecorous 
tales may be narrated without sullying the spirit of reverence 
which attaches to the regnant gods. 2 

Such is the raven among tribes on the North-Pacific 
coast. In the central plains of North-America we find as 

1 Ibiden p. 694—702. 

2 Hartley Burr Alexander , North-American Mythology p. 259. 

256 Jan de Vries, The problem of Loki FFC 110 

the culture-hero of the Menomini Manabus, who has a hu¬ 
man appearance. About him the same kind of tales are 
told . 1 He is the robber of fire and of tobacco and he survives 
the deluge by climbing on the top of a tree, from whence he 
despatches the beaver and the musk-rat to dive after land. 
But he is also the hero of a Jonah-story, and in a running 
contest he is tricked by an animal into losing his intestines. 
He imitates the magic arts of ethers by which he only makes 
things worse: by cutting slices of flesh from his wife’s back 
he hurts her severely and when he rubs her with earth it 
does not help a bit. 

I do not want to multiply the examples. I only wish to 
show what kind of stories usually are told about these super¬ 
natural beings, acting the role of culture-hero as well as 
trickster. Sometimes they have human, but more often 
animal forms. In the latter case the most frequent animal 
representations are the raven and the coyote; the raven on 
the Pacific Coast, the coyote among the Rocky-Mountain 
Indians. The human forms prevail in the Central and East¬ 
ern parts of the United States; here we hear not only of 
Manabus, but also of Gdooskap among the Penobscot and 
the Malecite Indians, Napi with the Blackfeet, the Old Man 
with the Crow-tribe, Inklonmi among the Assiniboins, 
Nihangan- among the Arapaho and several others. I may the 
more easily refrain from adducing examples about these 
supernatural beings as they are treated at full length in an 
excellent monograph of A. van Deursen, to which I bid the 
reader resort, if he wishes more information about this sub¬ 
ject . 2 

There is no difference between these conceptions whe- 

1 Skinner-Satterlee, Folklore of the Menomini-Indians in Anthro¬ 
pological Papers of the American Museum of Natural History XIII 
(New York 1915) p. 246 sqq. 

2 Der Heilbringer, Eine ethnologische Studie iiber den Heil- 
bringer bei den nordamerikanischen Indianern (Groningen 1931). 

FFG 110 Chapter XII, Loki as a religious phenomenon 


ther relating to a human being or an animal. In the folk¬ 
tales of many peoples the same stories about persons who 
make a fool of others or who are outwitted themselves, 
are told indiscriminately about clever persons, giants or 
animals. Hence Manabus is in this respect not different 
from the Coyote. But there may be slight variations in the 
character attributed to him; in one place he may be more 
dignified, in another however a more depraved type of 

The picture, drawn of the Coyote by Alexander, may 
be taken as an example of the other kindred types. He says 
then x : In multitude of stories he is represented as contempt¬ 
ible — deceitfull, greedy, bestial, with an erotic mania that 
leads him even to incest, often outwitted by the animals 
whom he endeavours to trick, without gratitude to those 
that help him; and yet, with all this, he is shown as a 
mighty magician, reducing the world to order and helping 
man with innumerable benefactions, perhaps less the result 
of his intention than the indirect outcome of his own efforts 
to satisfy his selfish appetite. It is impossible to regard such 
a being as a divinity, even among those tribes who make him 
the great demiurge; it is equally out of the question to regard 
him as a hero, for his character abuses even savage morals. 
In general he resembles the Devil of Mediaeval lore more 
than perhaps any other being —, the same combination of 
craft and selfishness, often defeating its own ends, of magic 
powers and supernatural alliances.» 

We are tempted to say: he is a coarse Loki, for the cha¬ 
racteristics attributed to the Coyote fit on the whole very 
well into the image of the Scandinavian trickster. Before 
making a comparison between Loki and the North-American 
corresponding mythical beings, it will be advisable to get a 
better insight into the structure of the latter. I 

1 L. c. p. 142. 



Jan de Vries, The problem of Loki 

FFC 110 

wish to quote Alexander \ who calles him »a being who is 
at once a demiurge, a magical transformer and a trickster, 
both clever and gullible. In some tribes the heroic character, 
in some the trickster nature predominates; others recognize 
a clear distinction between the myths in which creative 
acts are ascribed to this being and the folk-tales or fictions, 
in which his generally discreditable adventures are narrated*). 

As a culture-hero he even often acts in the still more 
elevated r61e of Creator. The feats told about him are: 

1. The setting in order of the shapeless first world and 
the conquest of its monstrous beings, who are usually 

2. The prime r61e in the theft of fire, the sun or daylight. 

3. The restoration of the world after the flood. 

4. The creation of mankind and the institution of the 
arts of life. 

If we compare these highly beneficial qualities of such a 
culture-hero, we are induced to ask: is it possible that he can 
be at the same time such a contemptible being, revelling in 
adventures of the most compromising kind? And the answer 
will be ready at hand: as he appears in the traditions of to¬ 
day, he seems to be the result of a long and complicated 
development, melting religious conceptions of utterly diffe¬ 
rent origin. Dr. Van Deursen who starts from the standpoint 
of the culture-hero, points out that he is an intermediary 
being between gods and men, that he is not the object of any 
veneration and that he is often represented as an animal. 
But there are also different historical elements in the tra¬ 
ditions about him, reminiscences of a former hero, chief, 
shaman or prophet. With regard to the curious mixture of 
the culture-hero and the trickster, Van Deursen contents 
himself with the following very prudent remark: The »Heil- 
bringer» appears often in the mixed r61e of an altruistic 

1 L. c. p. 298. 


FFC 110 Chapter XII, Loki as a religious phenomenon 

culture-hero, of a shaman and of a trickster; at one moment 
he acts as a helper of mankind, at another he sinks down to 
degradating acts. 1 

But how are we to understand the astonishing fact that 
a low-minded trickster and a beneficiant culture-hero can 
dwell together in one and the same personage? It is quite 
true, that the tradition, as it is now told, shows him as a 
figure that exercises a great influence upon all kind of 
tales and jokes. He is, as Van Deursen styles it, a »Konzen- 
trationsfigur»; we might say the funnel into which the scat¬ 
tered elements of popular lore are gathered. There are, 
indeed, several motives, current in almost all parts of the 
globe, that form a part of the cycle of the culture-hero only 
with the North-American Indians, whilst elsewhere they are 
told of animals or human beings, that have no particular 
relation to religious representations. It is obvious, as Prof. 
C. C. Uhlenbeck has remarked 2 , that in many respects these 
oral traditions have an unmistakable literary character. 
The heaping up of funny stories about tricksters and out¬ 
witted numskulls on the North-American »Heilbringer» 
shows that people are on confidential terms with him; he 
is their equal and their companion in all the trifles of every 
day life and we may compare him in this respect — and 
with all restrictions due to a comparison in a field like 
this — with the fox of European, the kantjil of Indonesian 
or Abu-Nuwas of Arabian tradition. Every new tale, in¬ 
vented or imported from elsewhere, every remarkable acci¬ 
dent or experience, will be transferred to Raven and Coyote 
and it is quite impossible to make use of this material for 
the study of the religious phenomenon itself. 

Still the fact remains, that he is at the same time a cul- 

1 L. c. p. 380. 

2 In a review of Van Deursen’s book in Tijdschrift van het 
Koninklijk Nederlandsch Aardrijkskundig Genootschap 2d Series 
XLVIII (1931), p. 1114. 


Jan de Vries, The problem of Loki 

FFC 110 

ture-hero and a trickster. The Indians consider their Mana- 
bus in this double aspect; still they do not regard him only 
in the light of the trickster-type. They speak of him rather 
as the hero and while they laugh heartily at his misadventu¬ 
res and enjoy the fun, they nevertheless specify that Mana- 
bus was foreordained by a power greater than himself to suf¬ 
fer and play pranks . 1 If we speak of a debasement of the 
original culture-hero into a rather discredited person, we are 
making a distinction between the conception of a lower and 
a higher order, that need not be the same as with the North- 
American aborigines. It is better not to call it a debasement 
but a displacement of the centre of gravity; the elements 
belonging to the type of the culture-hero are not liable to 
be increased: the world and its inhabitants have been crea¬ 
ted, cultural goods have been acquired, magical science has 
been revealed to mankind. On the other hand people’s 
fancy about the possible successes or discomfitures of a 
trickster is wellnigh inexhaustible and this stock of tradi¬ 
tions may be augmented continually. This is, however, of 
only secondary importance, for it is an undeniable fact, 
that from the very beginning the culture-hero as well as 
the trickster are present in the conception of this super¬ 
natural being. 

It may seem much more »logical», when the contradictory 
parts of this complex conception are divided over two distinct 
persons. With the Iroquois tribes a clear dualism exists 
between the twin-children of Ataentsic one is good and 
creates all profitable things, whilst the other thwarts him 
as much as he can. In the creation-myths one makes all the 
useful animals, the other however the nuisable and danger¬ 
ous creatures as serpents, wolves, bears and panthers. This, 
however, is a conception to be found among many other 

1 The last sentence quoted from Skinner and Satterlee, Menomini 
Folklore in the Anthropological Papers of Americ. Museum for Na¬ 
tural History XIII, 3 (1915) p. 236. 

FFC 110 Chapter XII, Loki as a religious phenomenon 261 

peoples also; in Melanesian mythology there is a good being 
Qat and a malevolent one Marawa. 1 In a previous chapter I 
have quoted the story, how Qat made human beings from a 
dracaena-tree; Marawa however tried to do the same, but 
he used the tavisoviso: he worked at them six days also and 
set them up, and beat the drum for them and gave them life 
as Qat had done for his. But when he saw them move he 
dug a pit, covered the bottom of it with cocoa-nut fronds, 
and buried his men and women in it for six days. Then when 
he scraped off the earth with his hands to view them, he 
found them all rotten and stinking; and this was the origin 
of death among men. Such is the story told in the Bank’s 
Islands; elsewhere the details may be different, but the 
underlying conception remains the same. On the Leper’s 
Island the benevolent being Tagaro creates edible fruit, 
human beings that walk erect and that dwell in houses; 
his opponent Suqe, however, makes fruits bitter and men 
in the shape of pigs who sleep in the trunks of sago-palms. 

Where a, clear dualism forms the principal feature of a 
religious system, we may expect that the opposition of good- 
and evilminded supernatural beings in the creation-myths 
is equally well known; in fact, European folk-lore has a pro¬ 
fusion of tales about God and the Devil creating the world 
and its inhabitants. I shall refrain from adducing examples, 
because they are accessible in the collection Natursagen of 
0. Dahnhardt. 

We may, perhaps, feel a certain inclination to consider 
this dualistic conception as the result of a later development, 
when the contradiction of such widely diverging characters 
in the same being became insupportable to a more enlight¬ 
ened mind. But I fully agree with Van Deursen, that both 
conceptions, the dualistic as well as the monistic, may be 
equally old and that it lies involved in the idea of the cul- 

1 B. H. Codrington , The Melanesians (Oxford 1891) p. 155 sqq. 


Jan de Vries, The problem of Loki 

FFG 110 

ture-hero itself, when his development follows one line or 
another. 1 Even in a dualistic system ;is that of the Bank’s 
Islands Qat is not exclusively a benevolent creator; Cod- 
rington says about him that he is »good-natured, only play¬ 
fully mischievous and that he thoroughly enjoys the exer¬ 
cise of his wonderful powers». And just as in America the 
Melanesian Qat is the central figure of a cycle of stories 
which vary in different parts of the islands of the Bank’s 
group. A certain ambivalence of character seems to be 
inherent with this type of the culture hero. 

There is assuredly no reason to regard the trickster as a 
mythological character of later origin than the culture-hero, 
as Lowie remarks 2 ; it is even possible that he is an older type 
of character in a given mythology than a properly so-called 
culture-hero, and then, as with the Assiniboins, the buffoon- 
type may prevail, although the sending out of the earth- 
diver birds, the theft of summer, the instruction in the kil¬ 
ling and skinning of the buffalo and the allotment of dances 
to various animals with orders to pass them on to mankind, 
may be attributed to him. In such a case the benefactions 
are not always given in an altruistic spirit, but only because 
the culture-hero wants to supply his own needs. 3 In the 
Raven-cycle he liberates the sun, not because he pities man¬ 
kind, but because he desires it. 

As a creator of all possible beings and things, nuisable as 
well as obnoxious he may be thought to participate of both, 
Primitive man did not imagine a creator of supreme good¬ 
ness, for he knew well enough that nature showed him a 
double face and that in man himself the same conflict was 
to be felt. The opposition of life and death was present in 
his mind; still he was persuaded that both life and death 
were in a constant interrelation and that in fact the mythical 

1 L. c. p. 373. 

2 R. Lowie, Journal of American Folklore XXII (1909) p. 431. 

8 F. Boas, Journal of American Folklore XXVII (.1914) p. 395. 

FFG 110 Chapter XII, Loki as a religious phenomenon 


spheres of both could not be divided. In many religions death 
is not regarded as the abnegation of life, but rather as a 
higher potential form of life. 

The dualistic conception of the universe prevails almost 
everywhere and reveals itself in many different forms. The 
opposition between life and death may be paralleled by 
that between heaven and earth or by that between the two 
sexes. It has its reflexion in social institutions based upon 
a dualistic principle as well as in typical products of mytholo¬ 
gical speculations. Twin-gods as the Dioskouroi or herma¬ 
phroditic beings as those from which the cosmos in many 
religions is supposed to evolve belong to the same religious 

This polarity is the essential feature of the culture-hero 
in several primitive religions. It does not matter, if it is 
expressed by opposing two supernatural beings belonging to 
different spheres of activity; it may be thought to exist as 
well in one single being. Then the religious phenomenon is 
particularly interesting, because the contrasting qualities 
cause a tension in the conception of the culture-hero, that 
may be of the utmost importance for its development on a 
higher level of civilisation. As an example the Greek Hermes 
may be mentioned, who originally belongs to this same cate¬ 
gory. 1 

The culture-hero is not a god, but only a supernatural 
being. Among his characteristics Van Deursen mentions that 
he is a mediator between the gods and men. He is not the 
object of special veneration. As a creator he is of minor im¬ 
portance in as much as he arranges the world and institutes 
a cosmic order. If he plays the r61e of a creator, it is com¬ 
monly done by moulding the world after the deluge. With 

1 Gf. J. P B. de Josselin de Jong , De oorsprong van den godde- 
lijken bedrieger, in mededeelingen der Kon. Akademie van weten- 
schappen te Amsterdam, Afd. Letterkunde deel 68. Serie B, N:o 1 
(Amsterdam 1929). 


FFC 110 

Jan de Vries, The problem of Loki 

this conception of North-American Indians the representa¬ 
tions of the Melanesians are in perfect accordance. When 
Qat is said to create he is adding only to the furniture of the 
world in which he was born, where there were already houses 
and canoes, weapons, ornaments, products of cultivated gar¬ 
dens and of such arts of life as the natives possessed when 
they were first visited by the Europeans. 1 But as a mediator 
between heaven and earth he participates of both and the 
contradictory elements of the cosmic duality have left their 
imprint on the conception of the culture-hero himself. 

This religious idea is well-nigh universal. The dualistic 
view on the world and its phenomena is so very self-evident, 
that it imposes itself upon the primitive mind. But the 
forms in which this fundamental conception is realised, are 
widely different. It depends upon the mental disposition of 
the people, upon its social institutions, upon the height of 
its civilisation and the influences that have given new impul¬ 
ses, which particular form the conception invests. When 1 a 
dissociation has taken place between two different beings, 
most commonly regarded as twins, the development leads 
to quite different results than where culture-hero and trick¬ 
ster are united in the same person. Hence we may expect to 
find at the last stage of historical evolution such very diffe¬ 
rent beings as e.g. a benevolent culture-hero nearly indistin¬ 
guishable from the Supreme God, a demiurge strongly in 
opposition to the heavenly powers, a messenger and media¬ 
tor between god and man, a kind of devil counteracting the 
creator in all possible ways, a mere trickster who is only 
the hero of comical stories and not to forget such a highly 
complicated figure as the Heilbringer of the North-American 

It is therefore in many cases exceedingly difficult to 
point out the culture-hero type as the original form of a 

1 Cf. Codrington 1. c. p. 155. 

FFC 110 Chapter XII, Loki as a religious phenomenon 265 

deity in a religion that has had a rich and variegated deve¬ 
lopment. When we find only the conception of a mere 
trickster left, obviously at this stage it would be impossible 
to attribute any religious meaning to it as a culture-hero. 
It may be very difficult to prove that he once has been of 
more importance than a hero of amusing tales. The fox 
of European folklore, on the other hand, is not an original 
culture-hero, because he executes tricks told elsewhere of 
such a supernatural being. It will be clear that it is extre¬ 
mely dangerous to prove or to reject the explanation of a 
deity belonging to a higher developed religious system as 
a culture-hero, only by confronting it with the well-testified 
types of ethnologic literature. The presence of some remar¬ 
kable elements does not prove identity nor does the absence 
of others disprove it. It is more or less a question of intui¬ 
tion, by which we must be led to the insight, that such a 
divinity by its explanation as an original culture-hero beco¬ 
mes more intelligible to us in as much as different elements, 
that formed a puzzling contrast, may be understood as 
the opposite aspects of one single religious conception. 

We may now return to the Scandinaviin Loki. But 
since the historical sources are scanty and Loki’s character 
is not easily to be defined, I think it advisable first to con¬ 
sider two other specimens of the culture-hero: the Creek 
divinities Prometheus and Hermes. 

3. Greek parallels of Loki 

A comparison between Loki and Prometheus lies near 
at hand. The bound titan fettered to a rock in the Cauca¬ 
sian mountains is a perfect counterpart of Loki punished in 
an analogous way for the murder of Baldr. In his famous 
treatise on the Ragnarok Axel Olrik lias shown the near 
relation between the myth of his punishment and the local 
tradition about the giant of Elbrus and he is of opinion 


Jan de Vries, The problem of Loki 

FFC 110 

that the Greek myth is a borrowing from the Caucasian 
legend. About the question how we have to explain Pro¬ 
metheus’ character as the robber of the fire, Olrik answers, 
that he as such a culture-hero certainly belongs to a primi¬ 
tive stratum of Greek religion; it is only his punishment in 
Mount Caucasus that ultimately goes back to the legend of 
the Elbrus-giant. Since the latter was a robber of water 
(in later development of the water of life) a connection be¬ 
tween him and Prometheus could easily have been made. 1 

It is not the place here to discuss Olrik’s hypothesis about 
the legends of the fettered giant. I am not quite convin¬ 
ced that it is of Caucasian origin, nor that Prometheus, 
Amiran and Loki go back to this tradition. There may 
have been an independant origin as well and the local legends 
of the Caucasian mountains may be themselves deeply 
influenced by literary or semi-literary traditions. For my 
purpose this is however of subordinate importance: because 
the punishment of Prometheus, however moving it may 
be in its unrivalled dramatic character, does not constitute 
an integral part of his myth, nor does it shed light on the 
meaning of this being. 

When Olrik calls him a culture-hero, he is evidently nea¬ 
rer to the truth than K. Bapp was, who Wrote a long and 
learned dissertation about Prometheus in Roscher’s Lexikon. 2 
To this scholar the Greek god is closely bound up with the 
element of fire; he must really have been an older form 
of Prometheus. The fickle nature of the element is expres¬ 
sed by the character of the Greek god, vacillating between 
good and evil. In the same philosophical but at the same 
time thoroughly intellectualistic way Prometheus as the 
creator of mankind is explained out of his supposed original 
meaning of a fire-god: »In the same way as fire, warmth, 

1 Axel Olrik, Danske Studier 1913, p. 107—121. 

2 Roscher’s Lexikon der griechischen und romischen Mythologie 
III, col. 3034 sqq. 

FFG 110 Chapter XII, Loki as a religious phenomenon 


soul, life were felt by the ingenuous realism of popular belief 
as well as by the older philosophers, as related, the god of 
fire and warmth could develop easily into a creator that 
bestows life to mankind)) 1 The dualistic character of Pro¬ 
metheus is, of course, explained by a later development: 
originally he had been a benefactor of mankind, but for the 
devout worshippers of Zeus the god of higher antiquity was 
seen in the unfavourable light of a rebel and an intrigant. 2 

Theoretical speculations of this kind are mere con¬ 
structions in which a former generation of mythologists 
liked to indulge. It is unnecessary to prove the umpro- 
bability of Bapp’s theory. To explain typical contrasts in 
a god’s character as the results of subsequent development 
is in most cases a proof, that the scholar has not been able 
to arrive at a conception of the divinity, in which both 
opposite qualities are reconciled as different aspects of one 
single religious representation. 3 

The real character of Prometheus is evident, as soon as we 
compare him with the North-American Manabus, Raven 
or Coyote. He is a very typical example of a culture-hero 
and the principal motives, Yan Deursen has pointed out in 
the myths about this supernatural being, may be found also 
in the Greek Prometheus. 

1. He is a kind of creator, but in fact more a renovator 
after a deluge has destroyed the world. According to 

1 Col. 3050—3051. 

2 A similar case is the myth of Salmoneus; first he was a divinised 
hero who caused the thunderstorm to rise, afterwards he was chast¬ 
ised by Zeus because he would have imitated the supreme god, cf. 
S. Reinach, Cultes, Mythes et Religions II, p. 166. 

3 I do not think it necessary, after all that has been said in the 
previous pages, to explain why I can not accept that Loki and 
Prometheus have been originally fire-demons. The reasons, addu¬ 
ced by L. von Schroeder in his Arische Religion are not convincing 
and especially his treatment of the Scandinavian Loki is quite in¬ 


Jan de Vries, The problem of Loki FFC 110 

different authors Prometheus is the creator of the 
first human couple, moulded by him out of water and 
earth (Apollodor and Plotin). When Zeus would de¬ 
stroy the men of the Bronze Age, Prometheus gave to 
his son Deucalion the advice to construct a chest and 
to embark in it with Pyrrha after having stored it with 

2. The culture-hero procures valuable things for mankind. 
As the robber of fire Prometheus is a typical form of 
the culture-hero and it is a serious mistake to consider 
him as a deification of the element of fire on account of 
his having stolen it from the gods. Sometimes it is 
said, that he has recovered it after it had been hidden 
by Zeus. Bapp translates this myth in the following 
way: »the beneficiant activities of fire remained intact 
even after the deluge». This is, at any rate, a very 
unpoetica.1 interpretation of the myth; I think it 
more likely that this isolated variant reflects the 
idea, that the culture-hero is not the original creator 
but only the remodeller of the world after the great 

3. He is an inventor of useful acquirements. Prometheus 
is sometimes considered as an educator of mankind; 
he has invented the alphabet, he is called a gramma¬ 
rian and dialectician, also an astrologer 1 Although 
these details certainly belong to a more sophisticated 
age, it is not unlikely that he has been supposed to 
be the inventor of such highly civilised accomplish¬ 
ments, because he already had been known as the 
god who taught to men e.g. magic properties. 

4. Of the utmost importance, however, is his dualistic 
character: at the same time benevolent and helpful, 
but also a deceiver and a rebel. His antagonism to Zeus 

1 Col. 3077. 

FFC llo Chapter XII, Loki as a religious phenomenon 269 

may be the cause that his hybris against the chief 
god has been aggravated into the most heinous act, 
Prometheus ever had committed. But this is probably 
the reflection of later philosophical interpretation, 
that wanted to emphasize the terrible conflict be¬ 
tween the superhuman hero and the gods, 
o. His position between gods and men may be compared 
with that of the culture-hero. In this connection we 
have precious testimony in one of Hesych’s glosses: 
Ithas; ho ton Titanon keruks Prometheus . 

I think no more proofs will be wanted. 1 The clever 
divinity, whose name means forethought, demiurge and 
herold of the pre-Olympian gods, is unmistakebly a typical 
form of the culture-hero. But he is thwarted in his career 
by the all-pervading superiority of Zeus. It is only his 
role as a robber of the celestial fire and as an actor in the 
drama of the deluge, that he has been remembered by later 

1 It is hardly necessary to add that I reject'the explanation offe¬ 
red by S. Reinach in his Cultes, Mvthes et Religions III, pp. 68—91. 
that Prometheus is the later anthropomorphic form of an original 
eagle, that brings fire to mankind. If he objects that other hypotheses 
are necessarily erroneous, if they do not explain the complete reli¬ 
gious phenomenon (p. 89 footnote 3), the same may be said with 
regard to his own theory. But, moreover, the idea that the fettered 
Prometheus might be compared with the eagle fixed to the front of 
the Greek temples and other buildings, seems to me utterly impos¬ 
sible, especially because the idea of punishment in this case must be 
considered as a later interpretation of the bound Prometheus, who 
obviously should have been represented fettered only because the 
eagle was fastened to the temple wall. I am unable to understand 
how a religious (or even mythological) conception may have origi¬ 
nated in this purely accidental way. In another place of this same 
book (II, pp. 171—172) he compares the torture of Prometheus 
with that of Tityos and supposes that they both go back to an image, 
representing the Titan, when he was struck down by Zeus and his 
corpse was left to the eagles; this is indeed a much more plausible 


Jan de Vries, The problem of Loki 

FFG 110 

generations. Struck by awe for the holiness of Zeus the 
Worshippers of this god considered Prometheus not in the 
first place as a benefactor of mankind, but as a reckless 
rebel against the majesty of the chief-god. His punishment 
was the result of this new interpretation; ungrateful man did 
not reject the precious gift Prometheus had bestowed upon 
him, but it loaded all its evil consequences upon the infor- 
tunate hero who was nailed to Mount Caucasus. 

In some respects Hermes is still a better representative 
of the dualistic culture-hero. The son of Zeus and Maia is 
placed in a near relation to the underworld as well as to 
heaven. He is a mediator between god and man, belonging to 
both worlds at the same time. He is moreover dualistic in 
more than one respect. His hermaphroditism is the expres¬ 
sion of his sexual ambiguity; but in the sphere of ethics 
we observe the same character because he is at the same time 
a benefactor and a deceiver. He is the inventor of the art of 
fire-making and of a new musical instrument; this clearly 
belongs to the nature of a culture-hero. But much more 
interesting is the fact that he is a typical trickster, many 
a time degrading himself to a clown, and moreover, like his 
American-Indian brother Coyote he is a very impudent 
seducer of women, being intricated in about thirty diffe¬ 
rent erotical adventures. Moreover his intimate relation to 
the powers of life and death is most likely mainly a speciali¬ 
sed form of his dualistic position between the celestial and the 
chthonic powers. Hence his r61e as a psychopompos, which 
at the same time proves his position as a kind of servant 
of the real gods; hence also his relation to the powers of 
fertility and lastly the sacrifices offered to him in the cult 
of the Hermes Chthonios. 

No more need be said about Hermes, whose character 
of a culture-hero and trickster is clearly exposed in the 
above mentioned paper of de Josselin de Jong. It becomes, 
however, evident from the fact that both this god and Pro- 

FFC 110 Chapter XII, Loki as a religious phenomenon 


metheus show the particular traits of this religious concep¬ 
tion, that at the end of historical development the results 
may be utterly different. Of course, we are not entitled to 
say that the Prometheus, as he appears in the classical age 
of Greece, is a culture-hero; he has become the hero of a 
well-defined mythological tale and he has lost accordingly 
many a feature, that probably at .an older stage of the reli¬ 
gious belief more sharply characterised him as the culture- 
hero. From the examples, adduced from ethnological litera¬ 
ture as well as from the Greek traditions, we may conclude 
that this kind of semidivine beings is very liable to become 
the central-point of epic cycles. But since the culture-hero 
himself shows a marvellous versability, these cycles may 
develop themselves in very different directions and even 
gradually lose all palpable contact with the original concep¬ 

4. Loki as a culture-hero 

The explanation of Loki as a culture-hero has been pro¬ 
posed by both von der Leyen and Axel Olrik. I have had 
the opportunity of mentioning it; now we will have to sup¬ 
port it with fresh, and as I believe convincing evidence. 
The proofs for this opinion have been hitherto those myths, 
where Loki indeed is told to have made or stolen such things 
as might be called culture-goods. He is the inventor of the 
fishing-net. He has stolen the Brisingamen and other use¬ 
ful objects. He has been caught as a salmon and in this 
fish a spark of the heavenly fire may have been concealed. 
H. Celander rejects this opinion and his arguments are 
worth considering. For as he says 1 , although Loki often 
steals different objects, these are never, properly speaking, 
such typical cultural acquirements as a culture-hero com- 

1 Lokes mytiska ursprung p. 116. 


Jan de Vries, The problem of Loki FFC 110 

monly is said to give to mankind. Elsewhere 1 he argues 
that the myth of the Brisingamen is an ordinary folk-tale 
and that the story of the capture of the salmon ultimately 
goes back to a Christian legend. Although I am not of the 
opinion, that Celander is right in this explanation, I readily 
agree that both myths are very frail supports for a hypo¬ 
thesis about Loki’s character as a culture-hero. We may add, 
that even the invention of the fishing-net is not sufficient to 
prove this characteristic. 

But I get the impression, that scholars have rather arbi- 
trarely restricted the religious phenomenon of the culture- 
hero to the acquirement of certain important cultural goods. 
Now we have seen that such a supernatural being in those 
religions, where we find it best preserved, is very composite 
and may present itself in widely different shapes. I have in¬ 
sisted upon the necessity to consider this phenomenon in 
its totality and especially to detect the central ideas that are 
connected with it. A comparison with cognate divinities in 
other religious systems may help us to make a clear distinc¬ 
tion between the fundamental character and the accessory 

I may begin with the discussion of the comparison, made 
by Haggarty Krappe 2 between Prometheus, Hephaistos, the 
Celtic Lug, the Scandinavian Loki and Lucifer. He mentions 
the following points: 

1. A divine or semi-divine being falls into disfavour with 
the powers that be (Prometheus, Hephaistos, Loki, 
Lucifer, Typhon), because. 

2. He is guilty of rebellion against the said all-wise and 
of course allgood powers (Prometheus, Hephaistos, 
Loki, Lucifer, Typhon). 

1 Danske Studier 1914, p. 84. 

2 The Science of Folklore (London 1930) p. 333. 

FFC HO Chapter XII, Loki as a religious phenomenon 


3. He is hurled down from the heavenly abodes (He- 
phaistos, Lucifer, Typhon). 

4. His fall explains the peculiar fact that he is limping 

•5. He is lame, though ;i different cause is given (Wav- 

6. He is identified with volcanic activities underground 
and earthquakes (Prometheus, Hephaistos, Loki, Ty¬ 

7. He is banished from the dwellings of the gods and 
fettered on a lonely rock (Prometheus, Loki, Lucifer). 

8. He is a fire-god and master-smith (Prometheus, 
Hephaistos, Lug, Loki, Wayland). 

9. He is fabled to be relieved of his chains in the last days 
when the end of the world is impending (Loki, Luci¬ 

10. He is a demiurgos, the inventor of many useful 
arts which he taught to man (Prometheus, Hephaistos, 
Lug, Loki, Wayland and Lucifer according to the 
tenets of certain gnostic sects). 

Such a, juxtaposition of different divine beings, characte¬ 
rised by common features, is exceedingly useful. But it seems 
to me that Haggarty Krappe is not very careful in the choice 
of the points of comparison and that as a consequence, the 
common character of these divinities does not become clear. 
He produces as different features such that really belong 
together as the succeeding stages of one single phenomenon 
(1, 2 and 3), or he puts as independant motives variants of 
the same item (3 and 7). Other peculiar traits are too spora¬ 
dical to deserve much attention in this connection (4 and 5 
are moreover variants of the same idea). Finally he adds 
such details, that possibly do not belong to the original 
conception and may be due to external influence (6, 7 and 
9 with regard to Loki). The result is a rather disorderly 



Jan de Vries, The problem of Loki 

FFC no 

combination of divinities that may belong partly to a similar 
religious representation and partly to quite a. different reli¬ 
gious sphere although they show some minor similarities. 

The most prominent feature of Loki is his mischievous 
character. He is a typical trickster, amusing himself with 
teasing the other gods and sometimes contriving to do serious 
damage. In most cases, however, he is obliged to repair it. 
Moreover he does not act so from pure malevolence, on the 
contrary he often has the best intentions, but it turns out 
bad. If in one myth he is a very harmless mischief-maker, 
in other cases he is going so far as to appear in the r61e of 
a sworn foe of the gods. It is probable, that in many respects 
the latter traditions are due to a later development; still 
we must not conclude that originally he only has been an 
inoffensive joker. Loki then is a culture-hero not in as much 
as he procures useful things for mankind, but especially in 
his predominant character of a trickster. 

It need not be said that this is not sufficient to prove the 
thesis that he indeed has been from the very beginning a 
culture-hero, for such a being can not be merely a trickster. 
But now those particularities that are found elsewhere as the 
constitutive elements of the culture-hero are of special in¬ 
terest. It even seems to me, that the fact that he is someti¬ 
mes the thief of divine objects (the Brisingamen, the apples 
of rejuvenation) or that he invents a useful art (the making 
of the fishing-net) gets an increased interest, because it 
fits into the conception of Loki as a culture-hero and a 

Like several of his colleagues he is in a very unstable 
position between the world of the gods and the demons. 
He in fact belongs to both, although the Scandinavian con¬ 
ception clearly lays stress on his divine character. But it 
can not be said that he is a full-blown god, he has more 
particularly the characteristics of a servant and a messenger, 
especially in those myths where he is the companion of 

FFG 110 Chapter XII, Loki as a religious phenomenon 275 

Thor. The strong resemblance with Hermes will at once 
become evident. 

These are all characteristic features, that belong to Loki 
as he appears in the numerous myths told about him, or in 
other words, that are essential to this divinity according 
to the conception, the last ages of paganism and the ensuing 
ages of literary elaboration had about him. But perhaps 
we may be so fortunate as to find in some few hidden corners 
the dim reminiscences about a culture-hero in a more dig¬ 
nified sense and above all as an element of religious belief. 
Now we are led to the opinion, that accidental features 
mentioned in our literary sources and sometimes very diffi¬ 
cult to understand, are really curious relics of a more original 
conception. The culture-hero belongs to different spheres 
of the universe and is often regarded as a dualistic being, 
even as bisexual. The same is told about Loki: he appears 
sometimes as a woman and even gives birth to children. 
I have discussed the possibility that this characteristic 
stood in some connection with the conception of a deity 
of fertility, although I did not venture to consider him 
exclusively as such. As soon as we explain him as a culture- 
hero this peculiar side of Loki becomes clear. For these 
supernatural beings are indeed supposed to promote fer¬ 
tility. Finally we have observed that Loki is closely related 
to the underworld (Hel, Fenrir); again we may point to 
the fact, that Hermes shows the same remarkable connec¬ 
tion with the realm of death. 

When so many details, that are at first sight incom¬ 
patible with each other, fit in so remarkably well into the 
conception of the culture-hero, we are entitled to accept 
this explanation. Where there was only contradiction and 
want of coherence, we now see a perfect unity. The scattered 
pieces of mythological lore have been readjusted into a 
pattern that does justice to the meaning of each single detail. 
Axel Olrik could only see in the different Scandinavian 

276 Jan de Vries, The problem of Loki FFC 110 

traditions about Loki the reflexion of three or four divine 
beings, that originally were quite independant. An expla¬ 
nation however, that succeeds in demonstrating the unity 
in the bewildering diversity of conceptions, must be pre¬ 
ferred because it is more simple and more convincing. 

Prom this point of view I wish once more to consider 
Loki’s relation to Othin. Axel Olrik has pointed out, that 
the myth about Othin stealing the mead of Suttung belongs 
to the same kind of stories as those current about Loki and 
he is of opinion that it goes ultimately back to a culture- 
hero myth about the origin of water . 1 Although this is by 
no means sure, the idea that Othin in traditions of this 
kind may be called a »stamfar», that is the divine ancestor, 
who is at the same time a culture-hero, seems to me fairly 
well founded. Then Othin and Loki are, at any rate, in this 
respect supernatural beings that belong to the same sphere 
of activities. The fostbrae&ralag between them may be 
the expression of this partial identity. 

Now in those primitive myths that told about a culture- 
hero, there was commonly one single person, Manabus or 
the Coyote. If there were two they were placed in some 
sort of contradictory position, the one being a benefactor, 
the other a malevolent being. We may adduce as an ex¬ 
ample the Iroquois cosmogony, where a heavenly woman 
is said to give birth to the twins Flint and Sapling. The 
latter one, better known as Yoskelia, is the real demiurge 
and earthshaper, he is too the spirit of life and of summer. 
But Flint, or Tawiscara, is an imitator and trickster, he 
makes malevolent beings and he is the spirit of wintry 
forces. Among their creative acts the most important is 
the moulding of man; this is the work of Sapling and when 
Flint tries to imitate him, he only succeeds in shaping 
monsters and this is why he is bannished to the underworld . 2 

1 Festskrift Feilberg p. 578. 

2 Hartley Burr Alexander . North American Mythology p. 36 sqq. 

FFC 110 Chapter XII, Loki as a religious phenomenon ill 

The antithesis shows itself in different ways: one thwarts 
or deceives the other; they are quarelling or they are active 
in different ethical spheres, one doing good and the other 
evil. Whilst the real culture-hero intends to bestow im¬ 
mortality on man, the other contrives in allotting death 
to him . 1 

Generally speaking, there is a certain correspondance 
with the divine pair Othin-Loki. For Othin is certainly 
rather the positive, and Loki the negative side of the culture- 
hero type. This is already evident in the tales about their 
robberies: Othin steals the mead from the giants and for 
the benefit of mankind, but Loki on the contrary steals 
IQunn from the gods to deliver her into the power of the 
demons. This is, indeed, exactly the contrast between the 
culture-hero and the trickster-deceiver. There are, how¬ 
ever, scarcely any indications that there has been a direct 
opposition between Othin and Loki, for Loki is never acting 
with intention to thwart Othin in one of his pursuits. 

Othin and Loki do not form a pair of quarrelling twins 
like Flint and Sapling. The reason may be that the cosmo¬ 
gonic myths, where Othin and Loki play a role, are nearly 
totally absent, perhaps because they have remained in a 
rudimentary state, more likely because they have been 
superseded by others. In fact, at the side of the myth about 
Ymir there is hardly any place left for the creative action 
of Othin. We know the creation-myth of Askr and Embla 
about him, to which we will return presently. In his Gylfa- 
ginning Snorri says about Alfabir: Hann smibabi himin ok 
jorb ok lopt ok alia, eign {teira; but it is very improbable 
that this represents a heathen tradition; it is at any rate 
more likely that Alfabir-Othin is here described according 
to the image of the Christian God. This becomes quite 

1 Gf. e.g. Gifford , Journal American Folklore XXXVI, p. 301 sqq; 
Van Deursen , Der Heilbringer p. 167, 177 and 317; F. Boas , Journa 
~ Am. Folklore XXX, p. 486—491. 

Jan de Vries, The problem of Loki 


FFG 110 

evident from tlie immediately following sentence, a curious 
mixture of heathen and Christian motives: hann gerdi 
manninn ok gaf honum ond pa, er lifa skal ok aldri tynask, 
pott likamr funi at moldu e5a brenni at osku; ok skolu 
allir menn lifa, peir er rett eru sidadir ok vera med honum 
sjalfum, par sem heitir Grimle, en v&ndir menn fara til Hel- 
jar ok paSan i Niflhel. 

According to the Scandinavian traditions Othin is not 
the creator of the universe, but only of mankind. And the 
myth telling about Askr and Embla, is presented in a form, 
that is difficult to reconcile with the type of the culture- 
hero. Olrik contends that the triad, to which this creation 
is ascribed, is unparalleled in Teutonic religion, but finds 
its exact counterpart in Celtic myths, where it is told e.g. 
that Lug, Dagde and Ogmios together go to the world of 
demons in order to recover the harp of Dagde. Then Olrik 
explains this curious similarity by the’hypothesis that this 
Celtic triad has been adopted by Germanic tribes living in 
the region of the Lower Rhine, and that it has been carried 
by a cultural current to Scandinavia, probably under the 
same circumstances as f the transmission of Othin himself 
would have taken place. 1 I have expressed my reasons, 
why I reject this latter view, in my paper on Othin. 2 But 
here a not less serious objection must be made: the creation 
myth mentions as the divine triad: Obinn, Hoenir and 
L65urr and I have in a former chapter (§ IV a) stated that 
we are not entitled to consider LoQurr as the same deity as 

The meagre material of our sources does not permit us 
to get any definite representation of Othin and Loki as a 
couple of divine beings, acting as demi-gods, that mould 
men and animals or procure useful arts for mankind. Othin 
is too markedly a divinity of high importance, to show any 

1 Festskrift Feilberg p. 578—579. 

2 FFComm., No. 94, p. 4:> sqq. 


FFG 110 Chapter XII, Loki as a religious phenomenon 

clear traces of an older form as culture-hero. In the tradi¬ 
tions about him we may detect two different layers: the 
original death-spirit and a divine creator. 1 But it is im¬ 
possible to make out whether the culture-hero has gradually 
developed into a creative god, e.g. in harmony with the rise 
of his importance in the Scandinavian pantheon, or whether 
Othin originally has been a divine creator. In my eyes the 
first alternative is the most probable. I may call to mind 
my previous remarks about Othin (or 05r), that one of his 
prominent characteristics has been fury or mental excite¬ 
ment, which is intimately connected with his intellectual 
quality as shown by the Old-Norse traditions. 2 This cha¬ 
racteristic, however, is closely bound up with all kind of 
magical practices; his name is identical with the Latin 
vales and the G-allec ouateis, both meaning »prophet». Such 
a name then may have denoted the primitive magician arid 
shamanic sorcerer and it is as such that the culture-hero 
not infrequently appears. We are inclined to accept this 
as the original meaning of 05r or Odinn. The demoniac 
character of Othin is especially the result of his belonging 
to the underworld and the world of the gods. This again 
connects him with several other culture-heroes. Concep¬ 
tions, belonging properly to the venerated ancestor and 
to the subterranean realm of death, may have merged into 
the primitive culture-hero, whose creative activity was 
mainly based on his magic properties. The later evolution 
into the supreme lord of the gods has, of course, totally 
obscured this original character of Othin. 

Loki was in many respects his counterpart. He worked 
on a lower plane, and what he did was not to the benefit 
of mankind. Perhaps the rise of Othin’s power had as 
consequence the gradual decline of Loki’s. His character 

1 Ibidem, p. 49. 

■ Ibidem, p. 45—46. 


Jan de Vries, The problem of Loki 

FFC llo 

shows, at any rate, several marked agreements with that 
of Othin. But in him the trickster type predominates. 

An explanation of the name Loki by means of a word 
logi »fire» or the verb lukan »to close» is irreconciliable with 
this conception. There is, however, a curious likeness to 
the name of the Celtic god Lug. And the connection of 
both is not confined to names only. The Romans have com¬ 
pared Lug with Mercury, and since Wodan was translated 
in the same way, there must have been some connection 
between their characters. According to Caesar Lug was an 
inventor of all possible arts and with this conception the 
Irish traditions about this god tally fairly well. For here 
also he was the inventor of the old heathen meetings at 
certain fixed times; he has introduced into Ireland several 
amusements, horse-races and the game of chess or at any 
rate an analogous play called fidchell. 1 When Lug arrives 
with the Tuatha De Danand before the battle of Mag- 
Tured, this god, called samildanach possessing many arts», 
gives an enumeration of his various abilities and we are 
again reminded of the Gaulish omnium inventor arlium. 

These are clearly qualities, that are characteristic of a 
culture-hero. The teaching of useful arts, the institution 
of religious rites are generally attributed to such divine 
powers. Lug was venerated as an important deity by the 
Gauls, as is proved not only by local names as Lugudunum, 
but also by the institution of the feast on the first of August 
in honour of the Emperor and the Majesty of the Empire, 
but in fact only continuing an old heathen festival on the 
same day in honour of Lug (the Irish lugnasad). There may 
have been, indeed, strong similarities between Lug and 
Wodan, as some scholars have pointed out . 2 This does not 

1 D-Arbois de Jubainville, Le cycle mythologique irlandais p. 138. 

2 The argument of Haggarty Krappe , Mythologie Universelle 
p. 222, that his name is derived from the word lugos »raven» and 
that this reminds one of the Teutonic ravengod, is very frail, because 

FFC I lo Chapter XII, Loki as a religious phenomenon 281 

exclude the possibility of a close relation between Lug and 
Loki, in as far as the latter belongs to the same category of 
divine beings and bears a remarkably corresponding name. 
The deplorable state of Celtic traditions about their heathen 
beliefs makes it an unpromising task to compare Scandi¬ 
navian deities with Celtic ones: when the former are not 
sufficiently clear to us, the latter will hardly help us in 
elucidating them. But when we confine ourselves within 
the borders of general observations, we may say with some 
confidence, that the analogous- character of Lug, Loki and 
Othin is a valuable indication for the existence of a prim¬ 
itive deity, belonging to the type of the culture-hero, among 
peoples so closely related as the Celts and the Teutons. The 
moment, however, when historical sources begin to shed 
some light upon the religious conceptions of both peoples, 
we observe that these deities have developed far from their 
original meaning and that the development went in very 
different directions. 

Hence if the similarity of the names Lug and Loki 1 
might be explained as a proof for an original identity of 
these deities, we can not expect to find any strong resem¬ 
blance between the Gaulish Lug and the Scandinavian Loki 
as they appear in the historical sources. But, of course, 
this does not prove that there could not have been one 
common original conception underlying both. Whenever 
we find the scattered fragments of a belief in a culture-hero 
among the variegated religious representations of later 
times, we may conclude from such erratic primeval boulders 
that they are indeed the last vestiges of by-gone ages, when 
a cruder and more primitive belief prevailed. 

the etymology of Clitophon: lougos »raven» is rather doubtful (cf. 
C. Dottin, La langue gauloise p. 268). 

1 The different stems (u-stem and consonant stem) are probably 
no serious objection against the identification of both names. 



Not every myth about Loki, preserved by the Old-Norse 
literary monuments is in accordance with the original 
character of this deity, as we have tried to establish it in 
the previous chapter. Loki the instigator of Baldr’s murder 
is quite a different personage from the friend of Othin or 
the companion of Thor. Whatever explanation of this god 
has been proposed, no scholar has arrived at such a concep¬ 
tion, that all the tales about him were made intelligible 
by it. Ideas of later ages, stories of foreign origin have been 
separated from the rather incoherent mass of traditions, 
till at last there remained a less numerous but more coherent 
material tallying with the theories advocated by the in¬ 

Now the later development of a religious conception 
is not simply an addition of new elements that originally 
do not belong to it. When a foreign tale is adopted and 
connected with an indigenous mythical figure, this is not 
done only because the tale was considered interesting and 
worth while recollecting. It was not connected arbitrarely 
with a mythical cycle, but there must have been reasons 
for doing so. Hence Christian traditions about Satan have 
not been the cause of Loki’s development into a devilish 
being, but they were connected with this god, because he 
possessed already certain qualities, that made him part¬ 
icularly fit as a heathen representative of the Christian 

FFC 110 Gh. XIII, Development of the conception of Loki 283 

devil. The evolution of a religious conception may be 
certainly influenced by foreign conceptions, but the fact 
that such an influence has taken place must again be ex¬ 
plained by motives lying in the original religious conception 

For a natural development, however, an indispensable 
condition is the persistence of a real belief in the deities, 
about whom the myths are told. This does not imply that 
the belief must be undisturbed, quite on the contrary, in 
many cases foreign influences may furnish new material 
for a richer growth of legends or myths and may even act 
as a stimulus to higher and nobler conceptions of the divine. 
But when a- belief is annihilated by a foreign creed or rele¬ 
gated to the state of superstition, circumstances are exceed¬ 
ingly unfavourable for a natural growth of a religious con¬ 
ception. This is what has taken place in Scandinavia. At 
a certain period tin- heathen belief is dethroned by Chris¬ 
tianity; its vestiges may have lingered on during many 
centuries in very remote places, the focusses of the cults 
have been dethroned and the intellectual part of the popula¬ 
tion, that is of the utmost importance for the elaboration 
of the mythological traditions, has abandoned the heathen 
belief. During the period of transition, perhaps one century 
or even more, there may have been a constant flowing in 
of Christian conceptions and legends into the decaying 
mythological system and as seems to be the case with the 
Voluspa, even give a higher meaning and a deeper sense 
to the pagan conceptions. But as soon as the vital force 
of heathendom is broken down, the mythological fancy 
must come to an end. 

Such, however, has not been the case in Iceland. By 
the particular character of the skaldic poetry there has 
arisen a new and unexpected growth of mythological tradi¬ 
tions. The use of kennings was firmly rooted in the heathen 
religious conceptions; in fact the majority of the poetical 


284 Jan de Vries, The problem of Loki FFC 110 

circumlocutions contain the name of a deity. Although 
during the first century after the conversion poets ns a rule 
were not inclined to adopt such kennings, because they 
reminded one too much of the ancestral creed, that Chris¬ 
tian priests had taught them to abhor and to depise, little 
by little they returned to the old custom and mythological 
kennings flourished as ever before. 

These kennings presuppose a certain knowledge of the 
heathen myths, not only amongst the poets themselves, 
but also with the public that listened to'their compositions. 
We must accept some kind of uninterrupted tradition that 
linked the modern renaissance with the heathen past. This 
tradition must have been purely literal and antiquarian; 
there have, of course, always been people interested in the 
old traditions, especially in Iceland, so extremely fond of 
old lore. The poems could not be understood unless people 
possessed the clue to the often very intricate kennings. 
This is the reason why the heathen traditions did not fall 
into utter oblivion as elsewhere in the Teutonic world: in 
the literary circles, where poets might learn the difficult 
skaldic art, the particulars of the heathen belief were ex¬ 
pounded in much the same way as classical mythology was 
taught in mediaeval convent-schools. When after a certain 
lapse of time the heathen traditions had become void of all 
religious meaning and were considered as curious tales from 
a distant past, there could be no danger anymore in using 
kennings with the names of the pagan gods; it might even 
appear as a new and attractive manner of poetical adorn¬ 
ment, giving some antiquarian flavour to the old-fashioned 
style of poetry. 

Then of course the mythological traditions were not 
simply collected and transmitted for the sake of explaining 
the old poems, but a new period of active interest for them 
set in. As soon as the poets made use of mythological ken¬ 
nings they would not restrain themselves from repeating 

FFC 110 Ch . XIII, Development of the conception of Loki 285 

the old ones, hut they had to invent new circumlocutions. 
Hence they had to reflect on the traditions themselves and 
in those indeed not so very rare places where the real mean¬ 
ing was obscure they had to seek for a plausible explanation. 
This means again a new stimulus to the learned preoccupa¬ 
tions of antiquarians, who tried to combine the scattered 
remains of heathen lore, to arrive at satisfactory inter¬ 
pretations, to readjust them into some kind of system. 
But as soon as this spirit is aroused, poetical fancy can not 
stand aloof; it too must participate in this work of revival 
of the old traditions. Taking their material as well from 
Christian legend as from classical stories, from popular tales 
and from their own imagination, the mythographers of this 
period extended the far too limited store of mythological 

Hence we may distinguish three successive periods in 
the transmission of heathen myths. In the ages before the 
conversion there has been a natural unconscious growth 
of the traditions about the divinities. It goes without saying 
that there have always been thinkers and poets, who freely 
handled the mythological traditions, even without believing 
in them. But they lived in a society where the majority 
actually was convinced of the reality of the deities and of 
the trustworthiness of the myths told about them. There¬ 
for the character of this period is that these myths must 
emanate from a living belief and consequently must be in 
harmony with the religious conceptions. But in as much 
as it was natural, the growth was wild and luxurious; there 
was but little disposition to systematisation and contra¬ 
dictory myths may have been current at the same time. 

Next comes the period immediately after the conver¬ 
sion. The. heathen traditions are preserved by those who 
have some interest in recollecting them. These are the 
learned antiquarians of this period, who have to furnish 
the knowledge for the right understanding of the old poetry. 

286 Jan de Vries, The problem of Loki FFG 110 

They may be compared with the filid of the insular Celts, 
who likewise handed down the old traditions to posterity. 
If the traditions are sufficient to explain the kennings, they 
have only to be preserved, but if this is not the case, some 
attempt may have been made to eke out the material; this 
may be done by combining different traditions and by fill¬ 
ing the gaps. If a kenning does not repose upon a known 
myth, it may lead to the invention of a tale apt to explain 
the kenning. 

At the end of the 12th century poets are using more 
freely mythological kennings. This means a heightened 
interest in religious traditions. If the activity unto this 
time had only been of a preservative kind, now it becomes 
also creative. Mythographers more freely handle the old 
material and they are especially intent upon a systematisa¬ 
tion of the very incoherent traditions. They, however, 
often misinterpret them, just like the poet who is »pillaging 
an antiquity of which he does not fully possess the secret». 
From the reciprocal action of poets and antiquarians new 
myths are added and old myths completely remodelled. 
Because nobody believes in them, they may be treated as 
purely literary traditions and it is unnecessary, even im¬ 
possible to take account of the real meaning of the old 
religious conceptions. For the modern scholar the activity 
of this period is a very serious hindrance to right under¬ 
standing; it is as if a screen were placed between the reliable 
traditions of the pagan time, by which they are in many 
cases utterly obscured. 

The ideal task of the modern scholar is to make a clear 
distinction between the products of these periods, especially 
between those of the first and the third. The latter he may 
disregard as to their meaning and in most cases also reject 
as quite spurious. They may contain some details that are 
original and interesting, because they have not been pre¬ 
served elsewhere; ordinarely they do not present such valu- 

FF C ll o Ch. XIII, Development of the conception of Loki 

able information. Tales of this period abound in mytho¬ 
logical names, that are mainly pure fancy. The thread 
Vartari by which Loki’s lips are sewed together belongs to 
this category, likewise the names Lording, Dromi and Gl-leip- 
nir for his fetters. 

The tales, however, belonging to the heathen time, are 
very precious for our investigation. They contain reliable 
material, they reflect original conceptions, they may con¬ 
sequently help us in finding the religious belief underlying 
those myths. When we are certain to base our conclusions 
upon this part of the extant material, we may be convinced 
that we build on a solid foundation. 

We get the impression that there are more spurious 
than reliable traditions. The majority of the myths told 
in the Snorra Edda are subject to this suspicion. Now here 
we find the most coherent information about them, written 
by a man who wishes to expound the old traditions before 
a public eager to get a comprehensive knowledge about 
them. Whereas the Eddie and skaldic poems in as far as 
they belong to the heathen time suppose the myths as known 
to everybody, Snorri wrote for those who wished to be 
taught. Hence his myths have the appearance of well 
ordered and intelligible stories, whilst the real reliable 
traditions are fragmentary and obscure. This makes it 
particularly difficult for us to abandon the guidance of 
Snorri and only to confine ourselves to the study of the old 
poems. Still we have met with several examples that Snorri’s 
combination proves to be quite arbitrary and consequently 
must be rejected. It goes without saying that this cha¬ 
racteristic of the prosaic Edda considerably diminishes its 
value as a first-rate source, and especially in those cases 
where we have to rely exclusively on its authority. The 
widely divergent opinions enounced by different scholars 
about the meaning of Old-Norse myths prove how ex¬ 
tremely difficult it is to arrive at definite conclusions in 

288 Jan de Vries, The problem of Loki F FC 11 0 

this field of study, owing to the incompleteness and obscu¬ 
rity of the extant material. 

Still it would be unwise to reject Snorri’s testimony 
altogether. This is impossible in those cases where he gives 
the only information about a myth. Moreover he may have 
had access to far better and richer sources of old lore than 
is possible for us, who live so many centuries afterwards. 
His interpretations, sometimes betraying the narrow¬ 
minded conceptions of mediaeval learning, may in other 
cases be founded on a better understanding of the heathen 
traditions, which may be ascribed to the fact that he was 
an Icelander himself and that he lived only a couple of 
centuries after the breakdown of paganism. We must bear 
in mind also that too great a scepticism necessarily deprives 
us of a considerable part of the material and consequently 
makes it well-nigh impossible to draw a vivid picture of the 
heathen belief. It may be preferable to involve a certain 
amount of spurious traditions in our investigations to pre¬ 
clude the wasting of the slightest piece of useful evidence. 
Hence I am inclined to place the largest part of the later 
material on the same level of trustworthiness as the most 
venerable traditions of pagan times. At any rate this may 
be justifiable, when we want to know the character of the 
divinity about whom the myths are told, because even 
later literary inventions will follow generally the same 
paths trodden by the heathen poets. 

When we try to collect the material about Loki that 
speaks in favour for our opinion, that he is a culture-hero 
and a trickster, we arrive at the conclusion that by far the 
greatest part of the myths of Loki are in perfect harmony 
with this conception. Only those stories that are generally 
regarded as due to foreign and Christian influence can not 
be reconciled with it. There are of course several contra¬ 
dictory tales about him. Can it be accepted e. g. that Loki 
in the same period may have been regarded as the friend 

FFC 110 Ch. XIII, Development of the conception of Loki 289 

of Othin, the servant of Thor, the father of Hel and of 
Sleipnir? I may then refer to the following passage of Max 
Muller 1 : »The poets of the Veda indulged freely in theogonic 
speculations, without being frightened by any contradic¬ 
tions. They knew of Indra as the greatest of gods, they 
knew of Agni as the god of gods, they knew of Varuna, as 
the ruler of all, hut they were by no means startled at the 
idea that their Indra had a mother, or that their Agni was 
born like a babe from the friction of two fire-sticks, or that 
Varuna and his brother Mitra were nursed in the lap of 

Moreover the myths themselves may be in full contra¬ 
diction to each other. This will be particularly the case 
when they are told about a divinity who possesses at the 
same time the wholly different characters of a trickster 
and a culture-hero. When we find fault with such a lack 
of congruity, the fault is ours. We are always craving for 
logical coherence and complete harmony; we always expect 
to find a well-ordered system and every time We only find 
fragmentary and contradictory traditions, we are prone 
to suppose an original harmonizing form that has been de¬ 
stroyed by ill-favoured traditions of later times. Observers, 
however, of primitive peoples have frequently insisted on 
this character of their myths and religious conceptions. 
Paul Radin calls special attention to the »freedom of thoughts 
among primitives and he says: »The same attitude is shown 
with regard to divergent versions of some of the more im¬ 
portant myths of the tribe, the sacred ones, those referring 
to the origin of the clans, of death, of future life. In one 
instance when I obtained a very markedly divergent version 
of the most sacred myth of the tribe, the informant, in reply 
to my question as to Why his version differed so much from 
the others, answered rather irritatingly »That is my way 

1 Vedic Hymns I (The Sacred Books of the East XXXII) p. 242. 



Jan de Vries, The problem of Loki FFC 110 

of telling the story. Others have different ways». That was 
all. No judgment was passed.# 1 

These preliminary remarks may pave the way for my 
attempt to give an explanation of the puzzling figure of 
Loki. He appears in the oldest sources as a very compound 
conception, although his general character is that of the 
trickster. On the whole he is not in direct opposition to 
the gods, for when he does make mischief to their damage 
he is compelled to repair it (cf. § X, i). The myths of J)jazi 
and I6unn, of the giant-builder, of Andvari and of the 
Brisingamen belong to this category, although I am by no 
means willing to call them all old and original. But here 
is a traditional conception that has been continued unto 
the latest times of myth-making. 

It would be as wrong to contend that those tales where 
Loki has a mischievous character, are to be considered as 
inventions of later ages. A trickster-type naturally is liable 
to transgress the line of demarcation between the admissible 
and the inadmissible. The trickster of primitive religions 
. shows the same proclivity to wicked acts as wm Icevisi Loki. 
The reason is obvious in those cases where he stands in con¬ 
trast to a twin-brother of benevolent character; then the 
duality necessarily pushes him on to the opposite side. The 
relation between Othin and Loki, although quite obscure 
in its real meaning, makes it probable that a development 
of the same kind has taken place with him. But it is not 
imperative to start from thi supposition because the free 
mythopoetic fancy very easily may add mischievous tricks 
to a divinity that originally was quite harmless. The tale 
of GeirroSr may illustrate this. Loki is a heinous traitor of 
the short-witted Thor, but nonetheless this character is 
not emphasized. He acted so because he was compelled 
to it by the giant who had made him a prisoner; this is 

1 Primitive man as a philosopher p. 56 f. 

FFC 110 Ch. XIII, Development of the conception of Loki 291 

evidently an excuse. Moreover he does not succeed at all, 
because Thor does not undertake the journey without magic 
objects that procure strength and because Thor conquers 
the giants. Finally he himself is described as a miserable 
being, clinging to the belt of the thundergod, when they 
pass through the river Vimur. 

In the tale about the cutting of Sif’s hair he acts as a 
naughty boy, and again he is forced to restore the damage 
he has done. Such a story shows clearly, that he sometimes 
may be the hero of very inoffensive and even childish tales. 
The same may be said about the myth of the wager between 
the dwarfs, immediately following the cutting of Sif’s hair: 
it belongs to the same style. The Lokasenna shows him in 
clear opposition to the gods, but his flyting contest in itself 
is not sufficient to regard him as a foe of the gods: in many 
a festival-hall of heathen Scandinavians the same shower 
of abusing and insulting remarks might be poured upon 
the heads of the guests by a sharp-tongued individual, 
whose selfpossession was slackened by the mead. The 
remark of the Hymiskviba that Loki had done damage to 
the he-goats of Thor is open to serious distrust: elsewhere 
it is told that f>jalfi was the real culprit and moreover the 
sentence about the godmolugir menn points to a later period 
of origin. 

The character of a trickster extolls the mental capacities 
of such a person at the cost of his bodily strength. I have 
shown (§ X, 3) that Loki’s companionship with Thor neces¬ 
sarily must have led to this conception. That is why Loki 
is described as a nimble being. The fact that he is said so 
often to be captured by the giants and to be overawed by 
the gods, is again an explanation for his bodily inferiority 
to both. We have no sufficient reason to call him an elfish 
being, only because he is described by Snorri as a person 
of small size. 

He sometimes acts as messenger, e.g. in the frymskviba. 


Jan de Vries. The problem of Loki 

FFC 110 

Although this poem in itself is certainly of late origin, this 
r61e is quite in agreement with the type of the culture-hero. 
This is proved by the Greek Hermes, who may furnish an 
interesting series of parallels with Loki. His relation to 
the underworld and to the powers of fertility, even his bi¬ 
sexual character, are explained by this side of his personality 
(cf. § X, 3 and 4). Traces of these conceptions, however, 
are only left in a few incidental remarks; the want of real 
myths may perhaps be explained by the greater predilec¬ 
tion for trickster stories which caused other sides of his 
personality to fall into oblivion. A harmless trickster easily 
becomes a clown. Hermes sometimes plays the r61e of a 
buffoon and Loki does quite the same in that scandalous 
tale, where, he succeeded in inducing Skadi to laugh. Those, 
who like Kaarle Krohn 1 explain this jest as an adaptation 
of a well-known popular tale ( Aarne-Thompson 571) are 
certainly nearer to the truth than those who seek for a con¬ 
cealed mythical meaning; nevertheless the motive of the 
clownish Loki in itself may belong to the oldest attainable 
form of this god. 

The same must be said of his many transformations. 
This is an essential characteristic of the trickster. Loki may 
appear as a salmon or a seal, as a mare or a falcon, as a fly 
or a flea; he may assume the outward forms of animals 
belonging to the three realms of nature. There is only one 
reason, why he in each case chooses a distinct shape and 
it must be admitted, that this reason is quite satisfactory: 
the shape he assumes is in perfect accordance with the aim 
he pursues. If he wants to conceal himself in the river, he 
becomes a salmon, but if he likes to fly through the air, he 
takes the shape of a falcon. As the- mother of Sleipnir he 
must be a mare and to disturb a person by a painful irrita¬ 
tion no better form could be adopted than that of an insect. 

1 Skandinavisk mytologi p. 177—178. 

FFC 110 Ch. XIII, Development of the conception of Loki 293 

HVnce there is no reason to seek for a special connec¬ 
tion between Loki and fire, or water, or even the fly-god 
Beelzebub to explain those myths where he appears as an 

If we may call the trickster-type the negative side of 
the culture-hero, we must ask, if there are any traces left 
of positive manifestations as such. The connection with 
Othin makes it fairly probable, that we should not put our 
expectations too high: in the pair Othin-Loki the former 
is chiefly the culture-hero, the latter the trickster. It is, 
however, not certain that in the original conception the 
duality has been clear-cut and there are indeed some indica¬ 
tions that both divinities are rather parallel than opposite 
phenomena. One might even be tempted to consider Othin 
as a kind of creator, Loki again as a divinity of a lower 
order, or rather as a culture-hero. The invention of the 
fishing-net is in itself not enough to stamp Loki as such, 
but in the light of our explanation some significance must 
be attributed to it, the more so because Loki is the only 
god about whom a myth of this kind is told. I am even of 
opinion that the myth, according to which the hammer of 
Thor owing to Loki’s wickedness has too short a handle, 
may be placed on a par with those primitive stories about 
a trickster who thwarts the goodwilled intentions of a god. 

This part of Loki’s character is, indeed,' obscured by 
the later profuse growth of the trickster-elements. The 
slight traces, however, left of an older and-higher concep¬ 
tion a re very precious, and Axel Olrik, who approached the 
enigmatic figure of Loki from quite a different side, was 
likewise induced to pay special attention to the so-called 
Othin-Loki and to explain him as a tribal-hero. 

A divinity as Loki has been, tends to a gradual debase¬ 
ment. The trickster type is much too human to remain 
in the same level as the gods. He acts too often as a con¬ 
temptible rascal or as a ridiculous clown to maintain his 

294 Jan de' Vries, The problem of Loki FFC 110 

divine character. The higher cultural niveau, to which the 
heathen Scandinavians raised themselves in the course of 
ages, necessarily implied a higher and nobler conception 
of the gods and compared with gods as Othin or Thor a 
being as Loki must lag behind. He was not deemed worthy 
to receive sacrifices; people did not pray to such a deity in 
hours of distress and sorrow. Whereas the names of lesser 
divinities like SkaQi or Horn are preserved in place-names, 
Loki has apparently never been the object of a public cult. 
Even in quite undisturbed pagan times Loki may have 
been degraded and as soon as a divinity is on this dangerous 
slope he must slide down deeper and deeper. During this 
period his character as a trickster remained virtually un¬ 
changed, but there clustered about him a greater mass of 
tales, where he acted as an impudent deceiver and buffoon. 
The small, sharp-witted but at the same time mischievous 
and cowardly Loki is the conception, that is in full harmony 
With such a development in malam partem. The ties that 
bound him to the realm of the underworld were tightened, 
but at the same time the unfavourable ideas of the world 
of the dead began to prevail. If Othin was gradually rising 
to the position of the Lord of heavenly Walhall, Loki on 
the contrary was degraded to Niflhel. He became the ally 
of the monsters that live outside the human world; as a 
lord of these infernal beings he might be called by a special 
term: tJtgarQaloki (cf. § X, 3). If the heathen Scandinavians 
had developed some ideas about a final break-down of the 
world when giants and demons would conquer the gods, 
Loki may have taken his place among these foes of the 
cosmic order. 

Here again it is difficult to make a clear distinction. 
For the-Christian influence has decidedly turned Loki into 
a heathen counterpart of Satan. Often, perhaps too often, 
the missionaries were seeking for points of contact between 
the Christian and the pagan religious representations and 

FFC 110 Ch. XIII, Development of the conception of Loki 295 

then a superficial similitude would suffice as an identifica¬ 
tion. The contrast of Othin and Loki might he seen in the 
exaggerated form of Alfa5ir and UtgarQaloki and then the 
comparison with God and Lucifer lies near at hand. 1 

Loki as a real devil may he due to Christian influence. 
His role in the myth of Baldr. in all probability the result 
of a later development, is an example of this new character. 
Likewise the tales about his being fettered in a subterranean 
cave, perhaps an imitation of legends about Lucifer, with 
whom Loki had an ominous resemblance of name. It is 
however of particular interest, that he has not received an 
important role in the drama of Ragnarok 2 , because we may 
infer from it that at the time of the poet of the Voluspa 
Loki was not yet debased to the heathen representative 
of the devil. It is not unlikely that there have been current 
even heathen myths about a punishment inflicted on Loki, 
because it was not to be expected that he could play his 
tricks on the gods without being chastised for it. The sewing 
of his lips by the dwarfs, his persecution in the river are at 
any rate other examples, that besides his being fettered, 
may prove that the idea of punishment also may have been 
current in heathen times. 

The belief, that he caused earth-quakes, is apparently 
of later origin. If, however, in heathen times people had 
experienced this natural phenomenon which they most 
likely had, then the trickster-god might have been thought 
responsible for it. The motive, however, is of too slight 

1 Of course I do not intend to sav that Othin himself has ever 
been put on' a par with the Christian God, but the notion of an 
Alfadir is very close to the idea of the supreme god sa er sdlina hefir 
ska pa i and here the difference with Christian conceptions is indeed 
very small (cf. I V H. Vogt. Vatnsdaelasaga p. LXXII—LX XIII). 

2 In Vafprudnismal st. 18 Surtr is called the principal opponent 
of tlie gods: Yigrfdr heitir vyllr, er finnaz vfgi at Surtr ok in svaso 

296 Jan de Vries, The problem of Loki FFG 110 

importance, to affect our view upon the real character of 
Loki, as I have developed it in the previous pages. 1 

The last period is that of the mythographers in the 12th 
and 13th centuries. Surely a not important amount of the 
myths told by Snorri and his contemporaries have been 
invented or at any rate have attained definite shape in these 
ages. But on the whole they seem not to have changed the 
idea of the god himself and evidently it is this they were not 
able to do. They had a well-fixed type of Loki in mind, the 
same type as the old traditions showed them and in as much 
as they were only poets or learned antiquarians, they did 
not feel any necessity to remodel the type itself. The char¬ 
acters of the heroes of Homer are also established firmly 
and Whatever later ages might add to their glory, a new 
conception of them lay beyond the scope of later poetical 

Perhaps we might find one example of an attempt to 
explain Loki in a new and original way. The eating-contest 
in the hall of G-eirroSr shows us Loki as the opponent of 
Logi, and as I have shown in § Y, 1 this implies that the 
author of this story supposed that Loki was also a kind of 
firegod. I do not know, if this author, who certainly did not 
invent the myth of Thor’s visit to the home of Geirrodr, 
but only remodelled and enlarged it, has explained Loki 
as a fire-god on the basis of a hap-hazard etymology or 
because the association of the devil and hell fire induced 
him to do so. At any rate the fire-nature of Loki appears 

2 The assertion of Miss Gras that this detail can not be recon¬ 
ciled with the explanation of Loki as an elfish spirit, is quite un¬ 
founded. Subterranean beings are often brought into connection 
with such phenomena and moreover in many tales an extraordinary 
force is attributed to them. Cf. the tale of the huldre that proves 
her strength by bending a horse-iron between her fingers or the 
legend of the dead king in the mountain who crushes an iron bar in 
stead of the finger he asked the human visitor to present to him. 

FFC 110 Ch. XIII, Development of the conception of Loki 207 

for the first time on the last stage of the literary traditions 
about him and there is no indication whatever that he has 
ever been connected with this element. This proves the 
impossibility of the hypothesis that Loki originally has been 
a fire-demon, but it proves at the same time that there are 
no indications.for Loki as a Scandinavian Prometheus: 
although a culture-hero, he seems not to have granted to 
mankind the valuable gift of fire. 

I am at the end of my investigation. The way has been 
difficult and wearisome; the reader may perhaps sometimes 
have despaired ever of reaching the goal. But many theories 
had to be disputed and many details had to be thoroughly 
investigated. The rock, on which we more than once were 
in danger of shipwreck, was the irreliability of a consider¬ 
able part of our material. Through the mind of a Christian 
a heathen conception must have been distorted. We may 
quote the words of the sage Mandanis to Onesikritos: It is 
impossible to explain philosophical doctrines through the 
medium of interpreters who know nothing of the subject. 
It is like asking water to flow pure through mud. 

But in spite of this serious drawback of our in many 
respects indispensable sources, we have arrived at a solu¬ 
tion of the problem by trying to find it, where we might 
expect that changes had been least at work. And this we 
found in the conception of the divinity, that has been un¬ 
altered during the different periods of historical develop¬ 
ment. The spirit animating the various acts of Loki must 
have been nearly the same. If we were able to find the 
central point of his personality, we might hope to find the 
clue of his enigmatic figure. I found it in his double char¬ 
acter of a culture-hero and a trickster and I consider it as 
a proof of no small importance that nearly all the traditions 
about Loki may be explained as the natural results of this 
original conception. 


Aegir 154, 171 -172, 206—207. 
Aestii 100. 

Aetiological legends 153. 

Agni 13, 289.- 
Aidan 75. 

Aimatar 191. 

Akaatar 191. 

A lpit; 99 I 

Alexander. H. B. 257—258. 

AJf 214. 

Alfadir 277, 295. 

Alfrigg 130,132. 

Ali 188—189. 

Aloen meri 152, 159—160. 
Alvaldi 38. 

Alvissmal 44, 67, 77. 

Alog 130—131. 

Amiran 266. 

Andvaranautr 41—44. 

Andvari 41—43, 61, 142—149, 
209, 290. 

Angrbo&a 75, 187—188, 198. 
Aphroditon 221. 

Ardshi Bordshi 34. 

Asgardr 65, 76—77/81. 

Asia Minor 163, 183. 

Asir 73. 78. 

Askr 28, 31, 33, 277—278. 
Atharva Veda 197. 

Attis 165. 

Atys 164. 

Aurora borealis 136—138. 

Baldr 23, 151, 163— 178, 185, 
199 217 295. 

Baldrs draumar 167 168, 179. 

Bapp, K. 266—268. 

Bardr Snsefellsass 214. 

Baruch 107. 

Bastian 87. 

Beelzebub 194, 233. 246, 293. 
Belit 194. 

Beowulf 139. 

Berlingr 130, 132. 

Bialfi 119. 

Biting louse 108—109, 113, 121. 
Bjarkamal 44, 91. 

Bjergmand 232—234, 247. 

Boer, R. G. 56. 

Brisingamen 26, 93, 95, 109,121, 
125—145. 150, 205, 271, 274, 

Brlsingr 136. 

Brisings gir&i 134—135. 

Brokkr 93, 94, 98, 176, 224. 
Brosinga mene 139. 

Bugge, S. 12, 14, 139, 198. 

Burr 28, 36. 

Byleistr 11,186—190, 193—195, 

Gahen, M. 193. 

Casuistic tales 34. 

Celander, H. 18—20, 27,101, 111, 
154, 207, 211, 214, 232—239, 

Chantepie de la Saussaye 15. 
Christian influence 23—24, 152, 
159, 182—183. 199, 214, 217, 
223, 278, 282—283, 288. 
Coyote 257, 259, 267, 270, 276 
Creation-myth 29—34. 
Culture-hero 16—18, 255—267. 

Dagde 278. 

Dagr 204. 

Dellingr 212. 

Deursen, A. van 256—265, 267. 
Dioscuric gods 168. 

Donar 55. 


List of Proper Najnes 

Draupnir 90—93. 

Dromi 98, 287. 

Dvalinn 130,132. 

Egill 85. 

Eillfr Gobrunarson 24, 56, 58, 
61—65, 186. 

Einarr Gilsson 44. 

Eiriksmal 196. 

Eisen, M. 122—123. 

Eldbjor 243—244. 

Eldir 173. • 

Embla 28, 31, 277—278. 
Emerson, E. R. 17. 

Esthonian tale 102—111, 120— 

124, 143—144. 

Eyvindr skaldaspillir 53, 196. 
Falcondress 85. 

Farbauti 11, 124, 186—190, 195, 
198, 205. 

Faroese ballad 45—49. 

Fenrir 10—11. 98, 186—190, 
195—197. 199, 208, 218, 223, 

Fettered demon 19,180,199, 265. 
Flmafeng 171—172. 

Finn 66, 70, 73, 75, 145—149. 
Finnish folktale 111—115. 
Finnish magic formulas 94,152— 
156, 190—194. 

Fire-spirit 228—229. 
Fjolsvinnsmal 211—212. 
Fjorgynn 204. 

Folkloristic material, value of 
5, 19, 118, 239—240. 

Folktales and myths 86—90. 
Forseti 204. 

Franangr 151, 171. 

Fredegar 43. 

Freki 197. 

Freyja 40, 66—67, 72, 78—85, 

125, 128—130. 

Freyr 52, 79. 

Frigg 40, 79, 170, 174, 176, 216. 

Geirrodr 24, 40, 56—65, 78, 
142, 145—149, 296. 

Geri 197. 

Gering 42, 51, 53. 

Geruthus 57. 

Giant-builder 65—82. 

Gimle 278. 

Gjalp 57, 62, 64. 

Gleipnir 98, 287. 

Glooskap 256. 

Gneip 57. 

Gobmolugr 97—98. 

Golther 13. 

Gosforth Gross 180. 

Graebner, F. 95. 

Gras, E. J. 21, 37—38, 41. 46, 
51—55, 62, 83, 92, 95—96, 
209, 296. 

Grendel 10. 

Grerr 130, 132. 

Gribr 63. 

Grimm, J. 10, 13, 43, 120. 
Grimnismal 93, 189, 193. 
Griiner-Nielsen 53. 

Gullinborsti 90—93. 

Gullveig 79. 

Gungnir 90, 92. 

Gustr 42. 

Gutasaga 119. 

Gylfaginning 28, 65, 82, 90. 96. 
Gondull 130—131. 

Hades 92. 

Hahn, J. G. von 92. 

Hakon the Good 183, 196. 
Hakonarmal 196. 

Haleygjatal 53. 

Hama 139. 

Hamel, A. G. van, 79, 153—154, 
171—176, 184. 

Haraldr hardradi 57. 
Harbar&sljdd 38, 41. . 

Hattatal 43. 

Haukr Valdfsarson 36, 53. 
Haustlong 23, 37, 39, 59. 134, 
136, ( 150, 186. 

Hebinn 130. 

Heilbringer 258—265. 

Heimdallr 23, 125—129, 133, 
138—141, 189, 204, 209. 

Hel 16, 19, 170, 186—189, 195, 
213—214, 223, 275, 289. 
Helblindi 184—190, 192—194. 
Hephaistos 10, 272—273. 
Herblindi 193. 

Hermann, P. 12. 

Hermes 263, 270—271, 275. 292.. 
H erodes 181. 

List of Proper Names 


Hervararsaga 39. 

Hesiod 33. I 

Hjabningavi'g 129, 138. | 

Hockert, R. 72. 78. 

Hoffory, J. 49. i 

Holmberg, U. 136—137. 
Hrafnass 38. 

Hreibmarr 43. 

Hrungnir 80—81, 118, 173, 235. ! 
Hdsdrapa 23, 125, 128—129, | 
139, 150, 186. 

Hustomten 230. 

Hymiskviba 18. 25, 84,100,113— ! 
115. 120, 195, 291. 

— St. 37—38, 96—97. I 

Hyndluljob 25. 75, 167, 187, 
198, 215—217, 220. 

Hobr 163, 167—168, 178—179, 

4 192. 

Hogni Halfdanarson 130. 

Horn 294. 

Hrenir 28—30. 35, 37— 47, 50, 
55, 146, 202, 204, 278. 

Ibunn 19, 40, 78,121,142—145, 
223, 290. 

Ildbjorg 243. 

Ilja 104—106, 109—110, 113. 
Ishtar 164. 

Isis 166. 

Islendinga drapa 36, 53. 

Ithas 269. 

Ivalda synir 90. 

Ivarr ljomi 129. 

Jakob Loj 232. 

J olios, A. 222, 224, 242. 

Jdnsson, F. 15, 42, 49, 59—60, 
95. 126—129, 201—202. 
Josselin, de Jong 270. 

Jumala 112 . 

Jummal 105. 

Jungner, H. 138, 222 . 

Jupiter 31. 

Jorb 61. 

Jormungandr 186, 190, 195 

‘ (cf. Midgardsserpent). 

Kali 221 . 

Kennings 38, 43—44, .59—61, 
91, 126, 133. 154, 178, 199, 

Kludde 22 , 209. 

Kock, E. A. 59—60, 64, 126—127 
Kormakr Ogmundarson 38. 

Kdu taat c 110 , 122 . 

Krappe, A. H. 65. 197, 272—273, 

Kreutzwald 102 — 111 , 120 , 122 . 
Krohn. K. 112—114, 152. 157— 
159, 181, 191, 216. 292. 
Kullebonden 232, 247. 

Kummer, H. 183. 

Kvasir 152—153, 158. 

Laakjen 246. 

Landtman, G. 234. 

Laufey 187—189, 191, 198, 205. 
Laxdoelasaga 126. 

Leland, G. G. 17. 

Lemminkainen 192, 216. 
Leviathan 115. 

Leyen, F. v. d. 17,• 34, 51, 83, 
86 — 88 , 271. 

Lijon 110 . 

Lind, E. H. 119. 

Lindqvist, I 60. 

Lockanat 155, 227, 236—237. 
Locke 229, 236—238 (cf. Loki). 
Locke Lewe 249. 

Locke Loye 249. 

Locknevi 52. 

Lodder 22 , 52—55, 209. 

| Lobkona 52. 

Loburr 28, 30, 35—37, 49—55, 

! 142, 209, 241—242, 249, 278. 

Loeke 53—55. 

Loga[>ore 22 , 51, 54—55. 

Logi 10 — 11 , 14, 17, 83, 160, 
162, 207, 241, 296 (cf. Loki). 
Lok 47—48 (cf. Loki). 

Lokasenna 15, 25, 81, 154, 167— 
178, 187, 194, 198, 206—207, 
215—216, 291. 

— st. 23: 213—215. 

— st. 27—28: 170, 174—178. 

— st. 50: 39. 

_ st. 54: 97 . 

Lokemand 226, 232, 247—248. 
(cf. Loki). 

Loki in Baldr-mvth 163—179. 

— cult of 203^-204. 

— as culture-hero 16,17, 20— 
21, 271—281, 293. 

302 List of Proper Names 

Loki relation to death 12. 16, 
19 211 213 

— and fertility 16, 166, 223. 

— and fire 4,10—12, 19. 83— 
84,151—162, 204—208. 228 
233, 244, 280. 

— as inventor 153—154, 236. 

— and lukan 10—15. 211, 
237—238, 280. 

— metamorphoses 66, 77, 85, 
155, 206. 

— and Othin 20—22. 28—55, 
142—144, 173, 201—202, 
205, 214, 220, 277,, 279, 

— in popular tradition 27, 

— and Ragnarok 179—186, 

— servant of Thor 16,18,21— 
22, 56—124,142—144.151, 

— as trickster 223—224, 248, 
253, 274, 292, 293. 

— as vatte 19, 205—207, 214, 

— and water 141, 156, 208— 
209 (cf. Locke, Loeke, Logi, 
Lok, Lokemand, Lokje, 

Lokje 226—227, 234—235, 246 
(cf. Loki) 

Lokka tattur 26, 45—48, 142— 
143, 146. 

Lokke 10, 45, 54—55, 151, 225— 
227, 232, 240—242, 247 (cf. 

Lokke Lagenson 249. 

Lokke Leyemand 249. 

Longinus 192. 

Loorits, O. 102, 104—107, 109— 
111, 113. 

Loptr 11. 49, 58, 205. 211, 213, 

Louhi 190. 

Loveatar 190—193. 

Lowie, R. 262. 

Loye Langbein 237—238. 

Lucifer 12, 14, 198, 211. 246, 
250, 272—273. 295. 

Lug 272—273, 278, 280—281. 

Lukas 232. 

Luna 221. 

Lundgren 119. 

Lyfte-stenen 116,118. 

Lsevateinn 211. 

Loeding 98, 287. 

MacCulloch 17, 224. 

Maia 270. 

Manabus 256—257, 260, 267, 276. 
Marawa 261. 

Mars 31. 

Matt ok megin 183—184. 
Menglob 212. 

Meyer,‘E. H. 4,12—13, 30,198. 
Meyer, R. M. 13. 

Michabous 155. 

Midgardsserpent 23, 58,114—115 
120—121, 188—189 (cf. Jor- 

Mjollnir 71, 90, r 92—93, 175. 
Mogk, E. 12, 14, 57, 79, 163, 
167—169, 204. 

Mogue, St. 75. 

Moses 117. 

Much, R. 11, 12, 135—136, 210. 
Muller, Max 289. 

Mythological tale 57, 65, 82. 

Naata 191. 

Nahanarvali 221. . * 

Nal 188—191, 198. 

Narfi 186, 188—189, 195, 198. 
Nari 188, 195, 205. 

Neckel, G. 163—165. 

Nerthus 221—222. 

Nibelungs 41—44, 46, 150. 
Ni&avellir 93. 

Niflhel 278, 294. 

Njorbr 79, 215. 

Nock 229. 

Nordal, S. 78. 

Nordendorf, brooch of 22, 51, 

Nott 204. 

Nuadu 147. 

Obinn cf. Othin. 

Obr 29, 73, 279. 

Ogmios 278. 

Ohrt, F. 48, 90, 94, 230. 

OJaf, saint, 68. 

Olaf Tryggvason 129—130. 

List of Proper Names 


Olafr Geirsta&aalfr 214. 

Olafr pai 23—24. 126. 

Olrik, A. 16—22, 27, 38, 45, 53, 
99—101, 111—112, 115—124, 
142, 180, 192—193, 208, 216. 
218, 225, 227—233, 240—241, 

, 247, 265, 271, 275—279, 293. 

Ori 212. 

Oschophoria 221. 

Osiris 166. 

Othin 10, 15, 20, 28—29, 79—81, 
130—132, 142—144, 150, 173, 
179, 181—182, 193—195, 201, 
214—216, 276—278, 294 (cf. 

Palmer, J. 31, 49. 

Paristaja poeg 107, 110. 

Perisokea 190—193. 

Persephone 82, 92. 

Petersen, N. M. 10. 

Philippson 48. 

Pikker 122. 

Pikne 122—123. 

Pipping, R. 66 : —67, 129, 136, 

Prometheus 10—11, 13, 16, 

166, 265—270, 272—273. 

Qat 34, 261—264. 

Radin, P. 32, 89, 289. 

Ragnarok 37,168,177,179—186, 
195, *197. 199, 207, 223, 295. 

Rampa 190—191. 

Ran 154, 171. 

Reginsmal 41—46, 50, 146, 150. 

Reinach, S. 269. 

Risin og Lokki 26. 

Routh, H. V. 32, 184. 

Rudbeck 116. 

Ruho 190—191. 

Rydberg 11. 

Rognir 59. 

Sacrifice to the fire 225—227, 
230—231, 244—245. 

— of a tooth 229—230, 243— 

Sahlgren, J. 52—53. 

Salmon 155—161. 

Saxo Grammaticus 57, 86, 163, 

Schiefner 120. 

Schoning, O. 15, 212. 

Schroder. F. R. 16, 53, 163, 166, 
222 . 

Schroder, L. von 208, 267. 

Schiick, H. 16. 

Schiitte, G. 43. 

Setala, E. N. 190—191. 

Seth 165—166. 

Sif 80—81, 90—93, 97, 143, 145- 
147, 224, 291. 

Sigrlami 39. 

Sigurd 42, 44, 219. 

Sigurdardrapa 38. 

Sigvn 166, 181, 188—189, 195, 

Sindri 93—94, 98. 

Singasteinn 125, 127. 129, 133, 
135, 139. 

Sinmara 211. 

Siva 221. 

Skabi 19, 37, 292, 294. 

Skaldskaparmal 125, 129. 

Ski'5blaSnir 90, 92—93. 

Sleipnir 66, 74—77, 144, 187, 
215—218, 289, 292. 

Snorra Edda 38—39, 61, 90,125. 

Snorri 25—28, 43, 57, 62—63, 
66 , 71—74, 77, 86, 93, 97—98, 
128—129, 133, 154, 177, 179, 
199, 287—288, 296. ' 

Spirits of the dead 230—231, 

Suqe 261. 

Suttungr 40. 

SvaSilfoeri 66, 74—75,187, 215— 

Svafnir 39. 

Svafrlami 39. 

Svipdagsmal 212. 

Sycfow, C. W. von 66—67, 71, 74 
77, 82—84, 212. 

Sijmons, B. 42. 

Saegiarn 211. 

Sorla[)attr 26, 128—130, 138, 
146, 150, 212, 235. 

Tagaro 261. 

Tawiscara 276. 

Thor 11, 23—24, 38—41, 56— 
124, 142—144, 156, 161, 173, 
175, 290—291 (cf. Loki). 


List of Proper Names 

Thorkillus 57. 

Titeliture 67—68, 74. 

Tmesis 126. 

Trickster 223—224, 242, 253— 
271, 290. 

Tuatha De Danand 280. 

Tuisto 220. 

Tune, runic stone of 129. 

Txa’msem 254—255. 

Typhon 272—273. 

Tyr 197. 

pjalfi' 18—19, 58—62, 82—84, 
99—101, 115, 117—120, 235. 

1>jazi 37—41, 45, 50, 145—150, 
199, 202, 290. 

Pjelvar 119. 

Pj6961fr Arnorsson 57. 

pjobolfr or Hvini 23, 37, 186. 

Porsdrapa 44, 56—58, 186. 

Prymr 235, 249. 

Prymskvi9a 18, 25, 40, 64, 80— 
81, 84—85, 100, 113, 120— 
121, 133—134, 136, 142, 146, 
187, 198, 201, 291. 

Pokk 170, 177, 217. 

Ukkonen 112, 123. 

Ulfr Uggason 23, 125, 128, 140, 

Ullr 189. 

UtgarSaloki 15. 82, 89, 144— 
149, 156, 162, 207, 213—214, 

Va5i 12. 

Vaf[>ru9nismal 44. 

Vagasker 125, 127, 129, 133. 140. 

Valil79, 189. 

Valshamr 40. 

Vana isa 122. 

Yanargandr 187. 

Yanataat 122. 

Vanir 29, 73, 78—79. 

Varr 212. 

Vartari 98. 176, 287. 

Ve 79. 

Velend 12, 273. 

Venus barbata 221. 

Vibofnir 211. 

Vikramacarita 34. 

Vili 79. 

Vimur 61, 63, 291. 

Vindhler 125. 

Yi tra 50. 

Vulcan 13. 

Voluspa 25, 66, 71, 76, 93, 166, 
*187, 283, 295. 

— st. 17—18: 28, 29—35, 58. 

— st. 22: 79. 

— st. 23—24: 80. 

— st. 25—26: 71—73, 77—78, 

— st. 32—34: 167—169. 

— st. 35: 168—169. 

Walhalla 81, 136. 

Wayland cf. Velend. 

Weinhold 91. 

Wesselski, A. 89, 115. 

Wilke, G. 13. 

Wisen, Th. 11, 208. 

Wod 47—48. 

Wodan 55, 280 (cf. Othin). 

Yama 221. 

Yima 221. 

Ymir 201, 221, 277. 

Ynglingasaga 79. 

Ynglingatal 23, 186. 

Yoskeha 276. 

Zeus 33, 267—270. 

Zrvan 221. 


Ch. I. Methodological Prolegomena . 3 

Ch. II. Critical Survey of the extant theories . 10 

Ch. III. Chronological Survey of the sources . 23 

Ch. IY. Loki and Othin . 28 

1. The myth of the creation of mankind .. 28 

2. The myth of fcjazi . 37 

3. The introduction of Reginsmal. 41 

4. The three gods in popular tradition. 45 

5. LoSurr and Loki . 49 

Ch. Y. Loki as the Companion of Thor . 56 

1. The tale of Geirrodr . 56 

2. The myth of the giant-builder . 65 

3. Other myths of Loki as companion of Thor 82 

4. Loki as Thor’s opponent. 90 

5. Loki as the servant of the thundergod .. 99 

a. Exposition of Olrik’s view . 99 

b. The Esthonian tale about the instru¬ 
ment of the thundergod .102 

c. The Finnish tale about the instrument 

of the thundergod . Ill 

d. Criticism of Olrik’s hypothesis .115 

Ch. VI. Loki and the Brisingamen .125 

1. Criticism of the literary sources .125 

2. The myth of the Brisingamen.133 



Ch. VII. The Character of Loki in Old-Norse Mythology 142 

1. Loki in his relation to Othin and Thor 142 

2. The double character of Loki .145 

Ch. VIII. Loki and the Element of Eire . 151 

1. Discussion of the sources . 151 

2. The fire-nature of Loki .161 

Ch. IX. Loki as the Enemy of the Gods . .163 

1. Loki in the myth of Balder’s death.163 

2. Loki and the Ragnarok . 179 

3. The family of Loki. 186 

4. The devilish Loki .198 

Ch. X. The Character of Loki .201 

1. Loki’s relation to the gods .201 

2. Loki’s relation to natural phenomena .. 204 

3. Loki’s relation to the underworld.210 

4. Loki’s bisexual character.215 

5. The ethical character of Loki .223 

Ch. XI. Loki and Modern Folk-lore .225 

1. Survey of the folkloristic material.225 

2. Olrik’s hypothesis about the fire-demon 


3. The hypothesis of Hilding Celander.233 

4. The value of the folkloristic material .. 239 

Ch. XII. Loki as a religious phenomenon .251 

1. The present state of affairs .251 

2. The trickster of primitive religions. 254' 

3. Greek parallels of Loki.265 

4. Loki as a culture-hero .271 

Ch. XIII. Development of the Conception of Loki .. 282 
List of Proper Names .299